抡锤者

sirwang

居然比被老特回复，那我就把前几天的简单LLM测试发一下数据，这是我前几天朋友圈发的：

第一手资料来了，vLLM 本地运行 Qwen3-8B 总占用32G, 权重占用8.8G/KV Cache占23，系统框架0.8G。 57.08 tokens/s，13.16 秒内生成了 751 个 token（包括思考过程和正式回复）。开启推理模式的情况下还能保持近 60 tokens/s，这表现还是相当让人满意的，这只是一块显卡。不到300W的功耗。和4090比起来还是相当给力的。现在用的FP8，改天试试FP16和多用户并发压榨测试，看能坚持得住几个人。新模型正在下载。个人感觉还是 qwen3.6-27b的会更帅一些。不接受反驳。

平台整体系统架构冗余度非常大。还有很深的潜力可以挖，当然，还没有正式进入生产环节。不知道同时运行3个视频生成流+一个本地大模型反推会是啥样的能耗表现……

开机...400W 只有两张卡运行-600瓦如果四张卡同时运行起来……看来我电费交少了……

sirwang

收到风： AMD / INTEL 下半年的工作室/工作站/个人的显卡发布态度：

ＡＭＤ：　主机集成性的带ＣＰＵ／内存共用的方式，直接发布就是主机，１６０Ｇ共用显存／内存起。
　　　　ＩＮＴＥＬ：　下半年发布用民用ＤＤＲ５颗粒的但大带宽（＞１.５Ｔ）的显卡，１６０Ｇ起。最高近５００Ｇ，我说的是显存！纯显存！

所以大家再等等也行。１０月左右我会拿到测试数据直接发出来，但大家拿到卡得在１２月了。　大家苦ＮＶＩＤＩＡ　久矣～～～

sirwang

https://github.com/intel/llm-scaler/tree/main

这是INTEL 官方公开的支持 B50/60/70 系列显卡的 comfyui 的docker 地址。他们还是做了不少适配的。下边有表：

https://github.com/intel/llm-scaler/tree/main#supported-models

Model Name FP16 Dynamic Online FP8 Dynamic Online Int4 MXFP4 Notes
openai/gpt-oss-20b
openai/gpt-oss-120b
deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
deepseek-ai/DeepSeek-R1-Distill-Llama-8B
deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
deepseek-ai/DeepSeek-R1-Distill-Llama-70B
deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
deepseek-ai/DeepSeek-V2-Lite export VLLM_MLA_DISABLE=1
deepseek-ai/deepseek-coder-33b-instruct
Qwen/Qwen3-8B
Qwen/Qwen3-14B
Qwen/Qwen3-32B
Qwen/Qwen3-30B-A3B
Qwen/Qwen3-235B-A22B
Qwen/Qwen3-Coder-30B-A3B-Instruct
Qwen/Qwen3-Coder-Next
Qwen/Qwen3.5-27B
Qwen/Qwen3.5-35B-A3B
Qwen/Qwen3.5-122B-A10B
Qwen/QwQ-32B
mistralai/Ministral-8B-Instruct-2410
mistralai/Mixtral-8x7B-Instruct-v0.1
meta-llama/Llama-3.1-8B
meta-llama/Llama-3.1-70B
baichuan-inc/Baichuan2-7B-Chat with chat_template
baichuan-inc/Baichuan2-13B-Chat with chat_template
THUDM/CodeGeex4-All-9B with chat_template
zai-org/GLM-4-9B-0414 use bfloat16
zai-org/GLM-4-32B-0414 use bfloat16
zai-org/GLM-4.5-Air
zai-org/GLM-4.7-Flash
ByteDance-Seed/Seed-OSS-36B-Instruct
miromind-ai/MiroThinker-v1.5-30B
tencent/Hunyuan-0.5B-Instruct follow the guide in here
tencent/Hunyuan-7B-Instruct follow the guide in here
Qwen/Qwen2-VL-7B-Instruct
Qwen/Qwen2.5-VL-7B-Instruct
Qwen/Qwen2.5-VL-32B-Instruct
Qwen/Qwen2.5-VL-72B-Instruct
Qwen/Qwen3-VL-4B-Instruct
Qwen/Qwen3-VL-8B-Instruct
Qwen/Qwen3-VL-30B-A3B-Instruct
openbmb/MiniCPM-V-2_6
openbmb/MiniCPM-V-4
openbmb/MiniCPM-V-4_5
OpenGVLab/InternVL2-8B
OpenGVLab/InternVL3-8B
OpenGVLab/InternVL3_5-8B
OpenGVLab/InternVL3_5-30B-A3B
rednote-hilab/dots.ocr
ByteDance-Seed/UI-TARS-7B-DPO
google/gemma-3-12b-it use bfloat16
google/gemma-3-27b-it use bfloat16
THUDM/GLM-4v-9B with --hf-overrides and chat_template
zai-org/GLM-4.1V-9B-Base
zai-org/GLM-4.1V-9B-Thinking
zai-org/Glyph
opendatalab/MinerU2.5-2509-1.2B
baidu/ERNIE-4.5-VL-28B-A3B-Thinking
zai-org/GLM-4.6V-Flash pip install transformers==5.0.0rc0 first
PaddlePaddle/PaddleOCR-VL follow the guide in here
deepseek-ai/DeepSeek-OCR
deepseek-ai/DeepSeek-OCR-2 There may be accuracy issues when using --quantization fp8
moonshotai/Kimi-VL-A3B-Thinking-2506
Qwen/Qwen2.5-Omni-7B
Qwen/Qwen3-Omni-30B-A3B-Instruct
openai/whisper-medium
openai/whisper-large-v3
Qwen/Qwen3-Embedding-8B
Qwen3-VL-Embedding-2B/8B follow the guide in here
BAAI/bge-m3
BAAI/bge-large-en-v1.5
Qwen/Qwen3-Reranker-8B
Qwen3-VL-Reranker-2B/8B follow the guide in here
BAAI/bge-reranker-large
BAAI/bge-reranker-v2-m3

sirwang

请把俩代码合一块。就可以了。

    {
      "id": 26,
      "type": "SaveImage",
      "pos": [
        1145.1501729206636,
        195.07454751045992
      ],
      "size": [
        267.9266338657344,
        433.279270302052
      ],
      "flags": {},
      "order": 21,
      "mode": 0,
      "inputs": [
        {
          "label": "图像组",
          "name": "images",
          "type": "IMAGE",
          "link": 29
        },
        {
          "label": "文件名前缀",
          "name": "filename_prefix",
          "type": "STRING",
          "widget": {
            "name": "filename_prefix"
          },
          "link": null
        }
      ],
      "outputs": [],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.6.0",
        "Node name for S&R": "SaveImage",
        "ue_properties": {
          "widget_ue_connectable": {},
          "input_ue_unconnectable": {},
          "version": "7.5.2"
        },
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        "ComfyUI"
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 11,
      "type": "PreviewImage",
      "pos": [
        1487.4226159090695,
        -382.86645688700804
      ],
      "size": [
        861.0544444444449,
        1199.7606666666666
      ],
      "flags": {},
      "order": 20,
      "mode": 0,
      "inputs": [
        {
          "label": "图像组",
          "name": "images",
          "type": "IMAGE",
          "link": 17
        }
      ],
      "outputs": [],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.6.0",
        "Node name for S&R": "PreviewImage",
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 25,
      "type": "Note",
      "pos": [
        2890.282933565041,
        -663.9119545246833
      ],
      "size": [
        421.42547299979583,
        1472.8439775316037
      ],
      "flags": {},
      "order": 3,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "properties": {
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        }
      },
      "widgets_values": [
        "正面视图低角度特写：<sks> front view low-angle shot close-up\n\n右前侧视图低角度特写：<sks> front-right quarter view low-angle shot close-up\n\n右侧视图低角度特写：<sks> right side view low-angle shot close-up\n\n右后侧视图低角度特写：<sks> back-right quarter view low-angle shot close-up\n\n背面视图低角度特写：<sks> back view low-angle shot close-up\n\n左后侧视图低角度特写：<sks> back-left quarter view low-angle shot close-up\n\n左侧视图低角度特写：<sks> left side view low-angle shot close-up\n\n左前侧视图低角度特写：<sks> front-left quarter view low-angle shot close-up\n\n正面视图平视特写：<sks> front view eye-level shot close-up\n\n右前侧视图平视特写：<sks> front-right quarter view eye-level shot close-up\n\n右侧视图平视特写：<sks> right side view eye-level shot close-up\n\n右后侧视图平视特写：<sks> back-right quarter view eye-level shot close-up\n\n背面视图平视特写：<sks> back view eye-level shot close-up\n\n左后侧视图平视特写：<sks> back-left quarter view eye-level shot close-up\n\n左侧视图平视特写：<sks> left side view eye-level shot close-up\n\n左前侧视图平视特写：<sks> front-left quarter view eye-level shot close-up\n\n正面视图高位拍摄特写：<sks> front view elevated shot close-up\n\n右前侧视图高位拍摄特写：<sks> front-right quarter view elevated shot close-up\n\n右侧视图高位拍摄特写：<sks> right side view elevated shot close-up\n\n右后侧视图高位拍摄特写：<sks> back-right quarter view elevated shot close-up\n\n背面视图高位拍摄特写：<sks> back view elevated shot close-up\n\n左后侧视图高位拍摄特写：<sks> back-left quarter view elevated shot close-up\n\n左侧视图高位拍摄特写：<sks> left side view elevated shot close-up\n\n左前侧视图高位拍摄特写：<sks> front-left quarter view elevated shot close-up\n\n正面视图高角度特写：<sks> front view high-angle shot close-up\n\n右前侧视图高角度特写：<sks> front-right quarter view high-angle shot close-up\n\n右侧视图高角度特写：<sks> right side view high-angle shot close-up\n\n右后侧视图高角度特写：<sks> back-right quarter view high-angle shot close-up\n\n背面视图高角度特写：<sks> back view high-angle shot close-up\n\n左后侧视图高角度特写：<sks> back-left quarter view high-angle shot close-up\n\n左侧视图高角度特写：<sks> left side view high-angle shot close-up\n\n左前侧视图高角度特写：<sks> front-left quarter view high-angle shot close-up\n\n正面视图低角度中景：<sks> front view low-angle shot medium shot\n\n右前侧视图低角度中景：<sks> front-right quarter view low-angle shot medium shot\n\n右侧视图低角度中景：<sks> right side view low-angle shot medium shot\n\n右后侧视图低角度中景：<sks> back-right quarter view low-angle shot medium shot\n\n背面视图低角度中景：<sks> back view low-angle shot medium shot\n\n左后侧视图低角度中景：<sks> back-left quarter view low-angle shot medium shot\n\n左侧视图低角度中景：<sks> left side view low-angle shot medium shot\n\n左前侧视图低角度中景：<sks> front-left quarter view low-angle shot medium shot\n\n正面视图平视中景：<sks> front view eye-level shot medium shot\n\n右前侧视图平视中景：<sks> front-right quarter view eye-level shot medium shot\n\n右侧视图平视中景：<sks> right side view eye-level shot medium shot\n\n右后侧视图平视中景：<sks> back-right quarter view eye-level shot medium shot\n\n背面视图平视中景：<sks> back view eye-level shot medium shot\n\n左后侧视图平视中景：<sks> back-left quarter view eye-level shot medium shot\n\n左侧视图平视中景：<sks> left side view eye-level shot medium shot\n\n左前侧视图平视中景：<sks> front-left quarter view eye-level shot medium shot\n\n正面视图高位拍摄中景：<sks> front view elevated shot medium shot\n\n右前侧视图高位拍摄中景：<sks> front-right quarter view elevated shot medium shot\n\n右侧视图高位拍摄中景：<sks> right side view elevated shot medium shot\n\n右后侧视图高位拍摄中景：<sks> back-right quarter view elevated shot medium shot\n\n背面视图高位拍摄中景：<sks> back view elevated shot medium shot\n\n左后侧视图高位拍摄中景：<sks> back-left quarter view elevated shot medium shot\n\n左侧视图高位拍摄中景：<sks> left side view elevated shot medium shot\n\n左前侧视图高位拍摄中景：<sks> front-left quarter view elevated shot medium shot\n\n正面视图高角度中景：<sks> front view high-angle shot medium shot\n\n右前侧视图高角度中景：<sks> front-right quarter view high-angle shot medium shot\n\n右侧视图高角度中景：<sks> right side view high-angle shot medium shot\n\n右后侧视图高角度中景：<sks> back-right quarter view high-angle shot medium shot\n\n背面视图高角度中景：<sks> back view high-angle shot medium shot\n\n左后侧视图高角度中景：<sks> back-left quarter view high-angle shot medium shot\n\n左侧视图高角度中景：<sks> left side view high-angle shot medium shot\n\n左前侧视图高角度中景：<sks> front-left quarter view high-angle shot medium shot\n\n正面视图低角度广角：<sks> front view low-angle shot wide shot\n\n右前侧视图低角度广角：<sks> front-right quarter view low-angle shot wide shot\n\n右侧视图低角度广角：<sks> right side view low-angle shot wide shot\n\n右后侧视图低角度广角：<sks> back-right quarter view low-angle shot wide shot\n\n背面视图低角度广角：<sks> back view low-angle shot wide shot\n\n左后侧视图低角度广角：<sks> back-left quarter view low-angle shot wide shot\n\n左侧视图低角度广角：<sks> left side view low-angle shot wide shot\n\n左前侧视图低角度广角：<sks> front-left quarter view low-angle shot wide shot"
      ],
      "color": "#223",
      "bgcolor": "#335"
    },
    {
      "id": 27,
      "type": "Note",
      "pos": [
        -1009.7430949133999,
        -667.218084639018
      ],
      "size": [
        3344.867777777771,
        172.79096969696934
      ],
      "flags": {},
      "order": 4,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "properties": {
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.8",
          "input_ue_unconnectable": {}
        }
      },
      "widgets_values": [
        "可以使用comfuyi-lumi-batcher 来跑各个角度的，一下子跑出去90条。\n在这个流里，直接换位置，也就是把位置选成节点19，之后把参数选成左边复制的提示信息就行. william"
      ],
      "color": "#232",
      "bgcolor": "#353"
    },
    {
      "id": 20,
      "type": "Note",
      "pos": [
        2392.2425113410113,
        -661.9303860055464
      ],
      "size": [
        466.503515625,
        1468.5173727560402
      ],
      "flags": {},
      "order": 5,
      "mode": 0,
      "inputs": [],
      "outputs": [],
      "title": "All prompt possible for the Lora Qwen image edit multiple angles",
      "properties": {
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        }
      },
      "widgets_values": [
        "<sks> front view low-angle shot close-up\n<sks> front-right quarter view low-angle shot close-up\n<sks> right side view low-angle shot close-up\n<sks> back-right quarter view low-angle shot close-up\n<sks> back view low-angle shot close-up\n<sks> back-left quarter view low-angle shot close-up\n<sks> left side view low-angle shot close-up\n<sks> front-left quarter view low-angle shot close-up\n<sks> front view eye-level shot close-up\n<sks> front-right quarter view eye-level shot close-up\n<sks> right side view eye-level shot close-up\n<sks> back-right quarter view eye-level shot close-up\n<sks> back view eye-level shot close-up\n<sks> back-left quarter view eye-level shot close-up\n<sks> left side view eye-level shot close-up\n<sks> front-left quarter view eye-level shot close-up\n<sks> front view elevated shot close-up\n<sks> front-right quarter view elevated shot close-up\n<sks> right side view elevated shot close-up\n<sks> back-right quarter view elevated shot close-up\n<sks> back view elevated shot close-up\n<sks> back-left quarter view elevated shot close-up\n<sks> left side view elevated shot close-up\n<sks> front-left quarter view elevated shot close-up\n<sks> front view high-angle shot close-up\n<sks> front-right quarter view high-angle shot close-up\n<sks> right side view high-angle shot close-up\n<sks> back-right quarter view high-angle shot close-up\n<sks> back view high-angle shot close-up\n<sks> back-left quarter view high-angle shot close-up\n<sks> left side view high-angle shot close-up\n<sks> front-left quarter view high-angle shot close-up\n<sks> front view low-angle shot medium shot\n<sks> front-right quarter view low-angle shot medium shot\n<sks> right side view low-angle shot medium shot\n<sks> back-right quarter view low-angle shot medium shot\n<sks> back view low-angle shot medium shot\n<sks> back-left quarter view low-angle shot medium shot\n<sks> left side view low-angle shot medium shot\n<sks> front-left quarter view low-angle shot medium shot\n<sks> front view eye-level shot medium shot\n<sks> front-right quarter view eye-level shot medium shot\n<sks> right side view eye-level shot medium shot\n<sks> back-right quarter view eye-level shot medium shot\n<sks> back view eye-level shot medium shot\n<sks> back-left quarter view eye-level shot medium shot\n<sks> left side view eye-level shot medium shot\n<sks> front-left quarter view eye-level shot medium shot\n<sks> front view elevated shot medium shot\n<sks> front-right quarter view elevated shot medium shot\n<sks> right side view elevated shot medium shot\n<sks> back-right quarter view elevated shot medium shot\n<sks> back view elevated shot medium shot\n<sks> back-left quarter view elevated shot medium shot\n<sks> left side view elevated shot medium shot\n<sks> front-left quarter view elevated shot medium shot\n<sks> front view high-angle shot medium shot\n<sks> front-right quarter view high-angle shot medium shot\n<sks> right side view high-angle shot medium shot\n<sks> back-right quarter view high-angle shot medium shot\n<sks> back view high-angle shot medium shot\n<sks> back-left quarter view high-angle shot medium shot\n<sks> left side view high-angle shot medium shot\n<sks> front-left quarter view high-angle shot medium shot\n<sks> front view low-angle shot wide shot\n<sks> front-right quarter view low-angle shot wide shot\n<sks> right side view low-angle shot wide shot\n<sks> back-right quarter view low-angle shot wide shot\n<sks> back view low-angle shot wide shot\n<sks> back-left quarter view low-angle shot wide shot\n<sks> left side view low-angle shot wide shot\n<sks> front-left quarter view low-angle shot wide shot\n<sks> front view eye-level shot wide shot\n<sks> front-right quarter view eye-level shot wide shot\n<sks> right side view eye-level shot wide shot\n<sks> back-right quarter view eye-level shot wide shot\n<sks> back view eye-level shot wide shot\n<sks> back-left quarter view eye-level shot wide shot\n<sks> left side view eye-level shot wide shot\n<sks> front-left quarter view eye-level shot wide shot\n<sks> front view elevated shot wide shot\n<sks> front-right quarter view elevated shot wide shot\n<sks> right side view elevated shot wide shot\n<sks> back-right quarter view elevated shot wide shot\n<sks> back view elevated shot wide shot\n<sks> back-left quarter view elevated shot wide shot\n<sks> left side view elevated shot wide shot\n<sks> front-left quarter view elevated shot wide shot\n<sks> front view high-angle shot wide shot\n<sks> front-right quarter view high-angle shot wide shot\n<sks> right side view high-angle shot wide shot\n<sks> back-right quarter view high-angle shot wide shot\n<sks> back view high-angle shot wide shot\n<sks> back-left quarter view high-angle shot wide shot\n<sks> left side view high-angle shot wide shot\n<sks> front-left quarter view high-angle shot wide shot"
      ],
      "color": "#232",
      "bgcolor": "#353"
    },
    {
      "id": 13,
      "type": "TextEncodeQwenImageEditPlus",
      "pos": [
        346.3903358490907,
        -56.08507473664949
      ],
      "size": [
        400.4109260819468,
        258.660770021565
      ],
      "flags": {},
      "order": 13,
      "mode": 0,
      "inputs": [
        {
          "label": "CLIP",
          "name": "clip",
          "type": "CLIP",
          "link": 18
        },
        {
          "label": "VAE",
          "name": "vae",
          "shape": 7,
          "type": "VAE",
          "link": 19
        },
        {
          "label": "图像1",
          "name": "image1",
          "shape": 7,
          "type": "IMAGE",
          "link": 20
        },
        {
          "label": "图像2",
          "name": "image2",
          "shape": 7,
          "type": "IMAGE",
          "link": null
        },
        {
          "label": "图像3",
          "name": "image3",
          "shape": 7,
          "type": "IMAGE",
          "link": null
        },
        {
          "label": "提示词",
          "name": "prompt",
          "type": "STRING",
          "widget": {
            "name": "prompt"
          },
          "link": 30
        }
      ],
      "outputs": [
        {
          "label": "条件",
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            3
          ]
        }
      ],
      "title": "TextEncodeQwenImageEditPlus (Positive)",
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "TextEncodeQwenImageEditPlus",
        "ue_properties": {
          "widget_ue_connectable": {
            "prompt": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65
      },
      "widgets_values": [
        "<sks> front view low-angle shot close-up"
      ],
      "color": "#232",
      "bgcolor": "#353"
    },
    {
      "id": 19,
      "type": "VNCCS_VisualPositionControl",
      "pos": [
        -135.64643889289317,
        113.93573315518134
      ],
      "size": [
        377.25527938354924,
        400.69737743530504
      ],
      "flags": {},
      "order": 6,
      "mode": 0,
      "inputs": [],
      "outputs": [
        {
          "name": "prompt",
          "type": "STRING",
          "links": [
            30
          ]
        }
      ],
      "properties": {
        "cnr_id": "vnccs-utils",
        "ver": "e8899e8fda5e72744198efecdc6f74f7d88a3b6a",
        "Node name for S&R": "VNCCS_VisualPositionControl",
        "ue_properties": {
          "widget_ue_connectable": {
            "camera_data": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        "{\"azimuth\":225,\"elevation\":-30,\"distance\":\"close-up\",\"include_trigger\":true}",
        ""
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 12,
      "type": "LoadImage",
      "pos": [
        -1045.7126666666695,
        -374.68955555555544
      ],
      "size": [
        850,
        1220
      ],
      "flags": {},
      "order": 7,
      "mode": 0,
      "inputs": [
        {
          "label": "图像",
          "name": "image",
          "type": "COMBO",
          "widget": {
            "name": "image"
          },
          "link": null
        },
        {
          "label": "上传",
          "name": "upload",
          "type": "IMAGEUPLOAD",
          "widget": {
            "name": "upload"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "图像",
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            10
          ]
        },
        {
          "label": "遮罩",
          "name": "MASK",
          "type": "MASK",
          "links": null
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "LoadImage",
        "ue_properties": {
          "widget_ue_connectable": {
            "image": true,
            "upload": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        },
        "#sdppp_variant": "default",
        "#sdppp_simple_content": "canvas",
        "#sdppp_simple_mask": "canvas",
        "#sdppp_simple_boundary": "canvas",
        "#sdppp_label": ""
      },
      "widgets_values": [
        "微信图片_20260515114607_5418_3.png",
        "image"
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    }
  ],
  "links": [
    [
      1,
      15,
      0,
      1,
      0,
      "MODEL"
    ],
    [
      2,
      5,
      0,
      2,
      0,
      "CONDITIONING"
    ],
    [
      3,
      13,
      0,
      3,
      0,
      "CONDITIONING"
    ],
    [
      4,
      1,
      0,
      4,
      0,
      "MODEL"
    ],
    [
      5,
      18,
      0,
      5,
      0,
      "CLIP"
    ],
    [
      6,
      14,
      0,
      5,
      1,
      "VAE"
    ],
    [
      7,
      8,
      0,
      5,
      2,
      "IMAGE"
    ],
    [
      8,
      8,
      0,
      7,
      0,
      "IMAGE"
    ],
    [
      9,
      14,
      0,
      7,
      1,
      "VAE"
    ],
    [
      10,
      12,
      0,
      8,
      0,
      "IMAGE"
    ],
    [
      11,
      4,
      0,
      9,
      0,
      "MODEL"
    ],
    [
      12,
      3,
      0,
      9,
      1,
      "CONDITIONING"
    ],
    [
      13,
      2,
      0,
      9,
      2,
      "CONDITIONING"
    ],
    [
      14,
      7,
      0,
      9,
      3,
      "LATENT"
    ],
    [
      15,
      9,
      0,
      10,
      0,
      "LATENT"
    ],
    [
      16,
      14,
      0,
      10,
      1,
      "VAE"
    ],
    [
      17,
      10,
      0,
      11,
      0,
      "IMAGE"
    ],
    [
      18,
      18,
      0,
      13,
      0,
      "CLIP"
    ],
    [
      19,
      14,
      0,
      13,
      1,
      "VAE"
    ],
    [
      20,
      8,
      0,
      13,
      2,
      "IMAGE"
    ],
    [
      22,
      16,
      0,
      15,
      0,
      "MODEL"
    ],
    [
      23,
      17,
      0,
      16,
      0,
      "MODEL"
    ],
    [
      29,
      10,
      0,
      26,
      0,
      "IMAGE"
    ],
    [
      30,
      19,
      0,
      13,
      5,
      "STRING"
    ]
  ],
  "groups": [],
  "config": {},
  "extra": {
    "workflowRendererVersion": "LG",
    "ue_links": [],
    "links_added_by_ue": [],
    "ds": {
      "scale": 0.40909090909091006,
      "offset": [
        1435.9227173859654,
        918.2021927921285
      ]
    },
    "frontendVersion": "1.43.18",
    "VHS_latentpreview": false,
    "VHS_latentpreviewrate": 0,
    "VHS_MetadataImage": true,
    "VHS_KeepIntermediate": true
  },
  "version": 0.4
}

sirwang

UP主很喜欢用刘悦的流，于是我就拿这个流来测一下B70吧。

上传图片 2000X3000 ，还是美女图。

我下载到的流，原装的流的提示词是破限的。无法公开展示所以我只改了几个字：

这是人家流里自带的时间计数器，我放大了一下移动了点儿位置。
视频wan2.2 ，480X854竖屏， 4 秒钟耗时 2分钟零7秒。没有任何其它优化。下边我会把流的代码贴出来。

群主说可以上传ZIP，看来我没权限，于是继续贴代码吧：

sirwang

前天开始，NV/AMD甚至INTEL的显卡都开始疯狂涨价。

但我个人建议大家可以淡定一些。纯粹个人建议。

因为我个人感觉，AI... 可能要崩掉了。好多个点儿都很....诡异。

看几个点儿吧。土耳其，亚马逊。

个人同意老特的那个观点--很快，就会有‘大船靠岸’。个人也觉得，因为NV要推动产品更新，所以有些硬件，虽然很不爽，但ZZZ确的也必须去更新的。原来我还以为国内的有些大厂会去批量购买这些卡，直到上次和字节的人沟通了一下才发现，低维护和损耗率以及稳定的可遇见性可控性才是他们所关注的。所以，这些大厂是不会买这些大船货的。

只是随便聊聊。个人建议，不够成参考。

PS. 和人沟通以后我才知道，现在全球才留内存颗粒/存储颗粒（硬盘）最大的居然不是我认为的亚马逊和谷歌，而是字节跳动，而且字节跳动的购买量居然是后边几个的...总量加起来还要多.... 太NB了。

sirwang

上一篇帖子是这个Inter B70，被要求重新开贴，所以有了这篇帖子：

事情是这样的：我想深度的测一下这卡的稳定性。如果长期去用，去批量跑任务，稳定性就很胆小。于是就有了这个操作：朋友让我帮忙处理一批图片，将图片 OCR 出来。图片都是2K+分辨率的。图片是一张大概有400-500行/10来列的表格。用QWEN3.6-27B去反推直接给OCR到excel表格里，我也想看看这卡的能耐咋样，之前有飞浆这些要钱的。也有github上开源的那些，但批量处理这么大的，我没用过。于是就写了代码，然后试了试这卡的能耐。只用了一张卡，从前天上午不到10点。到刚才。我截图也就是10分钟之前告诉我OK了。代码如下：

import base64
import os
import glob
import asyncio
import aiohttp
from io import BytesIO
from PIL import Image


API_URL = "http://localhost:8091/v1/chat/completions"

IMAGE_DIR = "./cb*.png"  
OUTPUT_CSV = "./cb_data_full_fixed.csv"

# B70 32G 显存并发数
CONCURRENCY = 6  

def encode_image_from_bytes(image_bytes):
    return base64.b64encode(image_bytes).decode('utf-8')

def slice_long_image(image_path, slice_height=1500):
    """
    核心修改：将超长图切片。
    slice_height=1500 像素大约包含 30-50 行数据。
    """
    img = Image.open(image_path)
    width, height = img.size
    slices = []
    
    for i in range(0, height, slice_height):
        # 截取切片区域 (left, upper, right, lower)
        box = (0, i, width, min(i + slice_height, height))
        slice_img = img.crop(box)
        
        # 将切片保存在内存中转为 base64
        buffered = BytesIO()
        slice_img.save(buffered, format="PNG") 
        slices.append(buffered.getvalue())
        
    return slices

async def fetch_and_process_slice(session, date_str, slice_base64, slice_index, file_lock):
    payload = {
        "model": "/model", 
        "messages": [
            {
                "role": "system",
                "content": "你是一个无情的数据提取机器。直接输出CSV，不要任何多余文字。"
            },
            {
                "role": "user",
                "content": [
                    {
                        "type": "text",
                        "text": "提取图片表格中所有可转债数据。请直接输出CSV格式。每行字段为：转债代码,转债名称,价格,涨幅,正股,正股价,溢价率。注意：不要包含表头，不要使用Markdown代码块（如 ```csv）。如果图片中没有完整数据行，请不要编造。"
                    },
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": f"data:image/png;base64,{slice_base64}"
                        }
                    }
                ]
            }
        ],
        "max_tokens": 4096,  
        "temperature": 0.0   
    }

    try:
        async with session.post(API_URL, json=payload) as response:
            if response.status != 200:
                print(f"⚠️ {date_str} (切片 {slice_index}) 请求失败")
                return

            res_json = await response.json()
            result = res_json['choices'][0]['message']['content'].strip()
            
            async with file_lock:
                with open(OUTPUT_CSV, "a", encoding="utf-8-sig") as f:
                    for line in result.split('\n'):
                        # 过滤掉可能的空行和重复生成的表头
                        if line.strip() and "," in line and "代码" not in line: 
                            f.write(f"{date_str},{line.strip()}\n")
            
    except Exception as e:
        print(f"❌ 处理 {date_str} (切片 {slice_index}) 发生异常: {e}")

async def main():
    image_list = sorted(glob.glob(IMAGE_DIR))
    if not image_list:
        print(f"❌ 错误：没有找到符合 {IMAGE_DIR} 的图片！")
        return

    print(f"🔥 找到 {len(image_list)} 张超长图，准备进行切片并高并发推断...")

    if not os.path.exists(OUTPUT_CSV):
        with open(OUTPUT_CSV, "w", encoding="utf-8-sig") as f:
            f.write("日期,转债代码,转债名称,价格,涨幅,正股,正股价,溢价率\n")

    semaphore = asyncio.Semaphore(CONCURRENCY)
    file_lock = asyncio.Lock()

    async def sem_task(session, date_str, slice_bytes, index):
        async with semaphore:
            slice_b64 = encode_image_from_bytes(slice_bytes)
            await fetch_and_process_slice(session, date_str, slice_b64, index, file_lock)

    timeout = aiohttp.ClientTimeout(total=None)
    async with aiohttp.ClientSession(timeout=timeout) as session:
        tasks = []
        for img_path in image_list:
            date_str = os.path.basename(img_path).replace("cb", "").replace(".jpg", "").replace(".png", "")
            
            # 对超长图进行切片
            slices_bytes = slice_long_image(img_path)
            print(f"✂️ {date_str} 被切分为 {len(slices_bytes)} 块，加入队列...")
            
            for index, slice_bytes in enumerate(slices_bytes):
                tasks.append(sem_task(session, date_str, slice_bytes, index))
        
        # 将所有切片任务并发执行
        await asyncio.gather(*tasks)

    print("🎉 全部长图切片处理完成！去检查数据量吧！")

if __name__ == "__main__":
    asyncio.run(main())

具体处理的图片不方便粘贴，但文件夹内的样子可以放一下。两个箭头一个是这个代码文件，一个是需要处理的图片有240多张。每一个图片都是1440宽，大概20000+像素高。

da5ea6d2-5fb1-474f-bc18-91ca83b8f844[1].jpeg

64ab71ee-809c-4fb4-83f4-bc54e5f49b25[1].jpeg
45c32027-bcfa-4d8e-a7b2-5c51850b8c1e[1].jpeg

显卡的温度和占用，只用看ID 3就行。：

37e86144-a85e-49b6-b6ab-d5c34b268ae8[1].jpeg

显卡的占用。不同的命令显示的有所区别。只用看ID 3就行。
c1746571-5620-47ab-ba5a-65a9e33987ab[1].jpeg

portainer 监控 docker 的截图
2330b65a-9a65-4106-b82f-83bd4f74f39d[1].jpeg

模型信息和docker运行的时间

0eab9876-94ed-4e85-8c44-1df8fdd42d35[1].jpeg

8c935b46-dd78-4719-ba7b-51f6561e3099[1].jpeg

可以看到全程这个GPU的占用率都在95%以上。时间用了16个小时。一直没停。结论是：这卡目前稳定性还是相当NB的，当然，也可能是和我的任务复杂程度有关系？现在是6个并发数，同时处理6个图片。这是第一批。第二批我会尝试加大并发处理量来再跑跑。

1dd7f10b-2d29-4696-9f06-e97e49bce01b[1].jpeg
忘了贴最终的数据量了。请原谅我的打码效果.... 哇哈哈哈
914b4289-9f63-4852-986b-93252852dcd5[1].jpeg

下边这张，是在linux上的截图，文件的创建时间是昨天上午的11.36，但在创建文件之前，代码已经运行了一个小时了，它得去把这200多个文件全部都截取成一个一个的小块才能读取数据OCR数据。所以文件时间就晚了一个小时。

OK。原贴贴完。以下是新内容。

sirwang

@t68823878 可以看到intel官方对于 AIGC的前景还是看好的，他们有一个团队去做这方面的技术支持，在comfyui的官方有了一个新的版本的comfyui去支持INTEL的卡。这是其1. 2 是在不同的模型适配上， wan/ltx2.3这些都OK了。有些LORA我还没试，可能有些弱，至于视频放大和一些用到cuda 和 nv gpu 的这些插件/custom node 就不要想了，虽然有些有 xpu 的支持，但性能还是有不少欠缺的。

他们官方为了解决入手门槛的问题，也紧急制作了docker 来让客户一键安装，但‘成也萧何败也萧何’ docker 的封闭性让 comfyui 的版本升级、pip配套环境的升级、git网络的使用都各种问题。

我已经建议他们将 comfyui 目录完全给映射到本地了。但现在的还是用起来极度别扭，一旦更新costom node 版本不对整个docker就崩溃，当然，这更多是我的问题。我尝试着去部署刘悦的这几个流，部署4天了。还没成功。等成功后我来给大家汇报它的效率以及1、2、3甚至4张卡的联合使用的效率。

同时也会根据老特他儿子的建议去跑一下Qwen3.6-27B ，他的建议是Qwen3.6-27B IQ4_K_M，我还没更多去看这几个的区别。但据他们官方说，这卡用 vllm 部署起来效率更高，请各位等我消息。

sirwang

OK。回来汇报来了。四张卡都驱起来了。机器有256G内存，一张卡分64G。前三张运行comfyui。后一张运行qwen3.6-27B. 测试大模型压力用的4并发。脚本和结果如下：

import urllib.request
import json
import concurrent.futures
import time

URL = "http://127.0.0.1:8091/v1/chat/completions"
HEADERS = {"Content-Type": "application/json"}
# 模拟长文本生成请求
DATA = {
    "model": "/model",
    "messages": [{"role": "user", "content": "请写一篇800字的科幻小说，描述人类第一次登陆木星的场景。"}],
    "max_tokens": 1000,
    "temperature": 0.8
}

def send_request(req_id):
    req = urllib.request.Request(URL, headers=HEADERS, data=json.dumps(DATA).encode('utf-8'))
    start_time = time.time()
    try:
        with urllib.request.urlopen(req) as response:
            res = json.loads(response.read().decode('utf-8'))
            tokens = res['usage']['completion_tokens']
            cost_time = time.time() - start_time
            print(f"请求 {req_id} 完成 | 耗时: {cost_time:.2f}s | 生成 Token: {tokens} | 速度: {tokens/cost_time:.2f} tokens/s")
    except Exception as e:
        print(f"请求 {req_id} 失败: {e}")

# 设置并发数，从 2 开始，逐步改成 4, 8, 16 试试极限
CONCURRENCY = 4 
print(f"--- 开始 vLLM 并发压测 | 并发数: {CONCURRENCY} ---")

with concurrent.futures.ThreadPoolExecutor(max_workers=CONCURRENCY) as executor:
    # 一次性发射 20 个请求排队
    executor.map(send_request, range(20))

以下是运行截图：

我画红线的是第4张卡运行vllm qwen3.6-27b的卡。

下边白底的这个是docker 的日志截图。

感觉还是相当稳的。工作室和个人用，够了。 comfyui 我去找个‘公平’的测试方法。或者大家有啥测试方法不？

sirwang

WAN2511 一张图生成96张图片的这个流，美女照，原图 2000X3000 ，成图的96张图 832x1248 。我简单看了下时间。从0.23分到0.43分，正好20秒。做为对比，我还有另外一台2080ti-22G魔改版，时间正好是快了9倍.... 这时间就差出大数来了。我还算比较满意。

2080ti-22G 同样的流，同样跑出来96个角度的图片：

2个半小时....

20分钟对2个半小时...

手里没有老特的4090-48G，否则铁定也要试一把~~~~

sirwang

为了方便大家测试，我把流共享出来：

{
  "id": "803716fc-9d9d-4d02-817c-855cdd6b4855",
  "revision": 0,
  "last_node_id": 27,
  "last_link_id": 30,
  "nodes": [
    {
      "id": 2,
      "type": "FluxKontextMultiReferenceLatentMethod",
      "pos": [
        797.1906256982577,
        12.11786329545784
      ],
      "size": [
        309.6734375,
        70
      ],
      "flags": {},
      "order": 15,
      "mode": 0,
      "inputs": [
        {
          "label": "条件",
          "name": "conditioning",
          "type": "CONDITIONING",
          "link": 2
        },
        {
          "label": "参考Latent方法",
          "name": "reference_latents_method",
          "type": "COMBO",
          "widget": {
            "name": "reference_latents_method"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "条件",
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            13
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "FluxKontextMultiReferenceLatentMethod",
        "ue_properties": {
          "widget_ue_connectable": {
            "reference_latents_method": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65
      },
      "widgets_values": [
        "index_timestep_zero"
      ],
      "color": "#222",
      "bgcolor": "#000"
    },
    {
      "id": 3,
      "type": "FluxKontextMultiReferenceLatentMethod",
      "pos": [
        797.1906256982577,
        -117.88213670454216
      ],
      "size": [
        309.6734375,
        70
      ],
      "flags": {},
      "order": 16,
      "mode": 0,
      "inputs": [
        {
          "label": "条件",
          "name": "conditioning",
          "type": "CONDITIONING",
          "link": 3
        },
        {
          "label": "参考Latent方法",
          "name": "reference_latents_method",
          "type": "COMBO",
          "widget": {
            "name": "reference_latents_method"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "条件",
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            12
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "FluxKontextMultiReferenceLatentMethod",
        "ue_properties": {
          "widget_ue_connectable": {
            "reference_latents_method": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65
      },
      "widgets_values": [
        "index_timestep_zero"
      ],
      "color": "#222",
      "bgcolor": "#000"
    },
    {
      "id": 4,
      "type": "CFGNorm",
      "pos": [
        807.1906256982578,
        -247.88213670454212
      ],
      "size": [
        270,
        68.33333333333334
      ],
      "flags": {},
      "order": 17,
      "mode": 0,
      "inputs": [
        {
          "label": "模型",
          "name": "model",
          "type": "MODEL",
          "link": 4
        },
        {
          "label": "强度",
          "name": "strength",
          "type": "FLOAT",
          "widget": {
            "name": "strength"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "修正后的模型",
          "name": "patched_model",
          "type": "MODEL",
          "links": [
            11
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "CFGNorm",
        "ue_properties": {
          "widget_ue_connectable": {
            "strength": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        1
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 17,
      "type": "UNETLoader",
      "pos": [
        -145.99967296522877,
        -365.28072730924794
      ],
      "size": [
        412.4183876274179,
        93.12632424766274
      ],
      "flags": {},
      "order": 0,
      "mode": 0,
      "inputs": [
        {
          "label": "UNET名称",
          "name": "unet_name",
          "type": "COMBO",
          "widget": {
            "name": "unet_name"
          },
          "link": null
        },
        {
          "label": "剪枝类型",
          "name": "weight_dtype",
          "type": "COMBO",
          "widget": {
            "name": "weight_dtype"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "模型",
          "name": "MODEL",
          "type": "MODEL",
          "slot_index": 0,
          "links": [
            23
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "UNETLoader",
        "ue_properties": {
          "widget_ue_connectable": {
            "unet_name": true,
            "weight_dtype": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "models": [
          {
            "name": "qwen_image_edit_2511_bf16.safetensors",
            "url": "https://huggingface.co/Comfy-Org/Qwen-Image-Edit_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_edit_2511_bf16.safetensors",
            "directory": "diffusion_models"
          }
        ],
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        "qwen_image_edit_2511_fp8mixed.safetensors",
        "default"
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 16,
      "type": "LoraLoaderModelOnly",
      "pos": [
        339.08373304272516,
        -365.4385152699498
      ],
      "size": [
        396.1328125,
        96.66666666666667
      ],
      "flags": {},
      "order": 8,
      "mode": 0,
      "inputs": [
        {
          "label": "模型",
          "name": "model",
          "type": "MODEL",
          "link": 23
        },
        {
          "label": "LoRA名称",
          "name": "lora_name",
          "type": "COMBO",
          "widget": {
            "name": "lora_name"
          },
          "link": null
        },
        {
          "label": "模型强度",
          "name": "strength_model",
          "type": "FLOAT",
          "widget": {
            "name": "strength_model"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "模型",
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            22
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "LoraLoaderModelOnly",
        "ue_properties": {
          "widget_ue_connectable": {
            "lora_name": true,
            "strength_model": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "models": [
          {
            "name": "Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors",
            "url": "https://huggingface.co/lightx2v/Qwen-Image-Edit-2511-Lightning/resolve/main/Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors",
            "directory": "loras"
          }
        ],
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65
      },
      "widgets_values": [
        "角度切换（VNCSS）-qwen-image-edit-2511-multiple-angles-lora.safetensors",
        1
      ],
      "color": "#432",
      "bgcolor": "#653"
    },
    {
      "id": 18,
      "type": "CLIPLoader",
      "pos": [
        -142.2511206355773,
        -201.03371333749183
      ],
      "size": [
        396.1328125,
        125
      ],
      "flags": {},
      "order": 1,
      "mode": 0,
      "inputs": [
        {
          "label": "CLIP名称",
          "name": "clip_name",
          "type": "COMBO",
          "widget": {
            "name": "clip_name"
          },
          "link": null
        },
        {
          "label": "类型",
          "name": "type",
          "type": "COMBO",
          "widget": {
            "name": "type"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "CLIP",
          "name": "CLIP",
          "type": "CLIP",
          "links": [
            5,
            18
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "CLIPLoader",
        "ue_properties": {
          "widget_ue_connectable": {
            "clip_name": true,
            "type": true,
            "device": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "models": [
          {
            "name": "qwen_2.5_vl_7b_fp8_scaled.safetensors",
            "url": "https://huggingface.co/Comfy-Org/HunyuanVideo_1.5_repackaged/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors",
            "directory": "text_encoders"
          }
        ],
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        "qwen_2.5_vl_7b_fp8_scaled.safetensors",
        "qwen_image",
        "default"
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 14,
      "type": "VAELoader",
      "pos": [
        -146.39059612263827,
        -17.514527994231976
      ],
      "size": [
        396.1328125,
        68.33333333333334
      ],
      "flags": {},
      "order": 2,
      "mode": 0,
      "inputs": [
        {
          "label": "vae名称",
          "name": "vae_name",
          "type": "COMBO",
          "widget": {
            "name": "vae_name"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "name": "VAE",
          "type": "VAE",
          "slot_index": 0,
          "links": [
            6,
            9,
            16,
            19
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "VAELoader",
        "ue_properties": {
          "widget_ue_connectable": {
            "vae_name": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "models": [
          {
            "name": "qwen_image_vae.safetensors",
            "url": "https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/vae/qwen_image_vae.safetensors",
            "directory": "vae"
          }
        ],
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        "qwen_image_vae.safetensors"
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 8,
      "type": "FluxKontextImageScale",
      "pos": [
        -133.49182202879047,
        560.4714628108627
      ],
      "size": [
        377.0828198681851,
        37.03860129882946
      ],
      "flags": {},
      "order": 9,
      "mode": 0,
      "inputs": [
        {
          "label": "图像",
          "name": "image",
          "type": "IMAGE",
          "link": 10
        }
      ],
      "outputs": [
        {
          "label": "图像",
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            7,
            8,
            20
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "FluxKontextImageScale",
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 15,
      "type": "LoraLoaderModelOnly",
      "pos": [
        347.0952420840532,
        -220.09228397794982
      ],
      "size": [
        396.1328125,
        96.66666666666667
      ],
      "flags": {},
      "order": 10,
      "mode": 0,
      "inputs": [
        {
          "label": "模型",
          "name": "model",
          "type": "MODEL",
          "link": 22
        },
        {
          "label": "LoRA名称",
          "name": "lora_name",
          "type": "COMBO",
          "widget": {
            "name": "lora_name"
          },
          "link": null
        },
        {
          "label": "模型强度",
          "name": "strength_model",
          "type": "FLOAT",
          "widget": {
            "name": "strength_model"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "模型",
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            1
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "LoraLoaderModelOnly",
        "ue_properties": {
          "widget_ue_connectable": {
            "lora_name": true,
            "strength_model": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "models": [
          {
            "name": "Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors",
            "url": "https://huggingface.co/lightx2v/Qwen-Image-Edit-2511-Lightning/resolve/main/Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors",
            "directory": "loras"
          }
        ],
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        "Qwen-Image-Edit-2511-Lightning-4steps-V1.0-fp32.safetensors",
        1
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 5,
      "type": "TextEncodeQwenImageEditPlus",
      "pos": [
        354.4459050818072,
        251.866217681473
      ],
      "size": [
        388.2640212658655,
        214.48069935058538
      ],
      "flags": {},
      "order": 11,
      "mode": 0,
      "inputs": [
        {
          "label": "CLIP",
          "name": "clip",
          "type": "CLIP",
          "link": 5
        },
        {
          "label": "VAE",
          "name": "vae",
          "shape": 7,
          "type": "VAE",
          "link": 6
        },
        {
          "label": "图像1",
          "name": "image1",
          "shape": 7,
          "type": "IMAGE",
          "link": 7
        },
        {
          "label": "图像2",
          "name": "image2",
          "shape": 7,
          "type": "IMAGE",
          "link": null
        },
        {
          "label": "图像3",
          "name": "image3",
          "shape": 7,
          "type": "IMAGE",
          "link": null
        },
        {
          "label": "提示词",
          "name": "prompt",
          "type": "STRING",
          "widget": {
            "name": "prompt"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "条件",
          "name": "CONDITIONING",
          "type": "CONDITIONING",
          "links": [
            2
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "TextEncodeQwenImageEditPlus",
        "ue_properties": {
          "widget_ue_connectable": {
            "prompt": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65
      },
      "widgets_values": [
        ""
      ],
      "color": "#322",
      "bgcolor": "#533"
    },
    {
      "id": 7,
      "type": "VAEEncode",
      "pos": [
        354.7124804040489,
        524.2547847261422
      ],
      "size": [
        384.81499821657553,
        60
      ],
      "flags": {},
      "order": 12,
      "mode": 0,
      "inputs": [
        {
          "label": "图像",
          "name": "pixels",
          "type": "IMAGE",
          "link": 8
        },
        {
          "label": "VAE",
          "name": "vae",
          "type": "VAE",
          "link": 9
        }
      ],
      "outputs": [
        {
          "label": "Latent",
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            14
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "VAEEncode",
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 1,
      "type": "ModelSamplingAuraFlow",
      "pos": [
        807.1906256982578,
        -357.88213670454206
      ],
      "size": [
        270,
        68.33333333333334
      ],
      "flags": {},
      "order": 14,
      "mode": 0,
      "inputs": [
        {
          "label": "模型",
          "name": "model",
          "type": "MODEL",
          "link": 1
        },
        {
          "label": "偏移",
          "name": "shift",
          "type": "FLOAT",
          "widget": {
            "name": "shift"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "模型",
          "name": "MODEL",
          "type": "MODEL",
          "links": [
            4
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "ModelSamplingAuraFlow",
        "ue_properties": {
          "widget_ue_connectable": {
            "shift": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        3.1
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 9,
      "type": "KSampler",
      "pos": [
        1140.647653458271,
        -367.5409128410178
      ],
      "size": [
        286.8991258117685,
        486
      ],
      "flags": {},
      "order": 18,
      "mode": 0,
      "inputs": [
        {
          "label": "模型",
          "name": "model",
          "type": "MODEL",
          "link": 11
        },
        {
          "label": "正面条件",
          "name": "positive",
          "type": "CONDITIONING",
          "link": 12
        },
        {
          "label": "负面条件",
          "name": "negative",
          "type": "CONDITIONING",
          "link": 13
        },
        {
          "label": "Latent",
          "name": "latent_image",
          "type": "LATENT",
          "link": 14
        },
        {
          "label": "随机种",
          "name": "seed",
          "type": "INT",
          "widget": {
            "name": "seed"
          },
          "link": null
        },
        {
          "label": "步数",
          "name": "steps",
          "type": "INT",
          "widget": {
            "name": "steps"
          },
          "link": null
        },
        {
          "label": "CFG",
          "name": "cfg",
          "type": "FLOAT",
          "widget": {
            "name": "cfg"
          },
          "link": null
        },
        {
          "label": "采样器",
          "name": "sampler_name",
          "type": "COMBO",
          "widget": {
            "name": "sampler_name"
          },
          "link": null
        },
        {
          "label": "调度器",
          "name": "scheduler",
          "type": "COMBO",
          "widget": {
            "name": "scheduler"
          },
          "link": null
        },
        {
          "label": "降噪",
          "name": "denoise",
          "type": "FLOAT",
          "widget": {
            "name": "denoise"
          },
          "link": null
        }
      ],
      "outputs": [
        {
          "label": "Latent",
          "name": "LATENT",
          "type": "LATENT",
          "links": [
            15
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "KSampler",
        "ue_properties": {
          "widget_ue_connectable": {
            "seed": true,
            "steps": true,
            "cfg": true,
            "sampler_name": true,
            "scheduler": true,
            "denoise": true
          },
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [
        173815537092855,
        "randomize",
        4,
        1,
        "euler",
        "simple",
        1
      ],
      "color": "#332922",
      "bgcolor": "#593930"
    },
    {
      "id": 10,
      "type": "VAEDecode",
      "pos": [
        835.9280787755798,
        137.81632044395135
      ],
      "size": [
        255.0718204517343,
        46
      ],
      "flags": {
        "collapsed": false
      },
      "order": 19,
      "mode": 0,
      "inputs": [
        {
          "label": "Latent",
          "name": "samples",
          "type": "LATENT",
          "link": 15
        },
        {
          "label": "VAE",
          "name": "vae",
          "type": "VAE",
          "link": 16
        }
      ],
      "outputs": [
        {
          "label": "图像",
          "name": "IMAGE",
          "type": "IMAGE",
          "slot_index": 0,
          "links": [
            17,
            29
          ]
        }
      ],
      "properties": {
        "cnr_id": "comfy-core",
        "ver": "0.5.1",
        "Node name for S&R": "VAEDecode",
        "ue_properties": {
          "widget_ue_connectable": {},
          "version": "7.5.2",
          "input_ue_unconnectable": {}
        },
        "enableTabs": false,
        "tabWidth": 65,
        "tabXOffset": 10,
        "hasSecondTab": false,
        "secondTabText": "Send Back",
        "secondTabOffset": 80,
        "secondTabWidth": 65,
        "ttNbgOverride": {
          "color": "#332922",
          "bgcolor": "#593930",
          "groupcolor": "#b06634"
        }
      },
      "widgets_values": [],
      "color": "#332922",
      "bgcolor": "#593930"
    },

貌似不行，那就再开一个贴再贴一半吧。我没有权限上传文件，只能贴代码了。

sirwang

模型用的官方原版的模型，没有量化。下载地址：https://huggingface.co/Qwen/Qwen3.6-27B/tree/main 一共 55.6G

sirwang

不知道，但从硬件发展上来看，这些厂商除了追求已经很难再提升的‘速度’就只剩下‘加大辅助’这一项了，也就是说只剩下加大显存这事了，哇哈哈哈。

现在的CPU /GPU 都已经过剩，再升级，提升的功耗比掏的钱老百姓更不会出钱了。

从这个角度来说，其实有可能他们也不知道加这么大显存，有啥‘完全可预料性’的解决方案吧？就目前来看，可能只有本地部署不压缩的模型了吧。

个人感觉。从长久来说，将来本地部署私人的大模型，起码对于我来说，会是个刚需，我现在所想做的方向，也是这个方向。随身小蜜，随身教练，随身陪护。就连现在的汽车和机器人，都在想着用用户端自己的大模型来处理这些远端的东西。所以说啊，未来可期。慢慢来。

sirwang

以下是新内容，前天的是处理的2024年全年的图片，有240多张，昨天处理了2026年的，只有 90张。为了更进一步的压榨这卡的能力，于是我把原代码中的6步并发，改为16个并发！我看是否它还可以抗得住，代码如下：

import base64
import os
import glob
import asyncio
import aiohttp
from io import BytesIO
from PIL import Image

 
API_URL = "http://localhost:8091/v1/chat/completions" 
IMAGE_DIR = "./cb*.png"  
OUTPUT_CSV = "./cb_data_full_fixed.csv"

# B70 32G 显存并发数
CONCURRENCY = 16  

def encode_image_from_bytes(image_bytes):
    return base64.b64encode(image_bytes).decode('utf-8')

def slice_long_image(image_path, slice_height=1500):
 
    img = Image.open(image_path)
    width, height = img.size
    slices = []
    
    for i in range(0, height, slice_height):
 
        box = (0, i, width, min(i + slice_height, height))
        slice_img = img.crop(box)
        
        buffered = BytesIO()
        slice_img.save(buffered, format="PNG")  
        slices.append(buffered.getvalue())
        
    return slices

async def fetch_and_process_slice(session, date_str, slice_base64, slice_index, file_lock):
    payload = {
        "model": "/model", 
        "messages": [
            {
                "role": "system",
                "content": "你是一个无情的数据提取机器。直接输出CSV，不要任何多余文字。"
            },
            {
                "role": "user",
                "content": [
                    {
                        "type": "text",
                        "text": "提取图片表格中所有可转债数据。请直接输出CSV格式。每行字段为：转债代码,转债名称,价格,涨幅,正股,正股价,溢价率。注意：不要包含表头，不要使用Markdown代码块（如 ```csv）。如果图片中没有完整数据行，请不要编造。"
                    },
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": f"data:image/png;base64,{slice_base64}"
                        }
                    }
                ]
            }
        ],
        "max_tokens": 4096,  
        "temperature": 0.0   
    }

    try:
        async with session.post(API_URL, json=payload) as response:
            if response.status != 200:
                print(f"⚠️ {date_str} (切片 {slice_index}) 请求失败")
                return

            res_json = await response.json()
            result = res_json['choices'][0]['message']['content'].strip()
            
            async with file_lock:
                with open(OUTPUT_CSV, "a", encoding="utf-8-sig") as f:
                    for line in result.split('\n'):
                        # 过滤掉可能的空行和重复生成的表头
                        if line.strip() and "," in line and "转债代码" not in line: 
                            f.write(f"{date_str},{line.strip()}\n")
            
    except Exception as e:
        print(f"❌ 处理 {date_str} (切片 {slice_index}) 发生异常: {e}")

async def main():
    image_list = sorted(glob.glob(IMAGE_DIR))
    if not image_list:
        print(f"❌ 错误：没有找到符合 {IMAGE_DIR} 的图片！")
        return

    print(f"🔥 找到 {len(image_list)} 张超长图，准备进行切片并高并发推断...")

    if not os.path.exists(OUTPUT_CSV):
        with open(OUTPUT_CSV, "w", encoding="utf-8-sig") as f:
            f.write("日期,转债代码,转债名称,价格,涨幅,正股,正股价,溢价率\n")

    semaphore = asyncio.Semaphore(CONCURRENCY)
    file_lock = asyncio.Lock()

    async def sem_task(session, date_str, slice_bytes, index):
        async with semaphore:
            slice_b64 = encode_image_from_bytes(slice_bytes)
            await fetch_and_process_slice(session, date_str, slice_b64, index, file_lock)

    timeout = aiohttp.ClientTimeout(total=None)
    async with aiohttp.ClientSession(timeout=timeout) as session:
        tasks = []
        for img_path in image_list:
            date_str = os.path.basename(img_path).replace("cb", "").replace(".jpg", "").replace(".png", "")
            
            # 对超长图进行切片
            slices_bytes = slice_long_image(img_path)
            print(f"✂️ {date_str} 被切分为 {len(slices_bytes)} 块，加入队列...")
            
            for index, slice_bytes in enumerate(slices_bytes):
                tasks.append(sem_task(session, date_str, slice_bytes, index))
        
        # 将所有切片任务并发执行
        await asyncio.gather(*tasks)

    print("🎉 全部长图切片处理完成！去检查数据量吧！")

if __name__ == "__main__":
    asyncio.run(main())

直接上图

这是结果，图片太少，不知道啥时候完成的，应该是昨天晚上半夜。

16并发时的显卡压力，频率到了2583，显卡瓦数到了228.还没有到顶，这卡到顶300，官方说是290，但我的确用到过瞬时300. 显存占用率 96.91% 。

这是在运行时的模型吞吐量。大概230-280 tokens/s。这超出了在开始测试时的180tokens/s 。有懂的可以告诉我为啥... 同样是 Avg Generation troughput 为啥在直接和它对话时在180/s 而现在却到了280多？是模型预热好了？费解。

不管咋说，26年的90张处理完毕，下一步计划是把并发增加到... 26？再试试25年的数据，25年有240多张图片。尽请期待测试结果。

sirwang

新发布了个400. 160G显存，啥也不缺了。

sirwang

手里有INTEL 的 B70PRO 显卡，新发布的 32G显存。
可以用comfyui，用 z-image 生图，会强过4090，但LTX/WAN上边，没办法720视频，适配的一塌糊涂。我都快没有信心去测试了。 comfyui也没办法更新。我正在调试。调试完之后第一时间来发报告。

sirwang

INTEL 下半年要发布的卡，PCB 裸图。显存从160G到差不多500G。这是要掀桌子的节奏。

sirwang

@sospda 地主家也没有余粮。 NV的经济大头在企业。INTEL的单卡如果再不激进就还是回到芯片里边吧。连自己看不上的AMD都搞不过，不丢人吗？

sirwang

看着就热血沸腾。我在想我的4卡B70 PRO 是不是也搞搞这个......

sirwang

Intel_Arc_Validation_Suite_v1.zip

初始版本。准备写的更详细些。

抡锤者

sirwang

帖子