<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Strata/单卡7900xtx 配置一例 (IQ3_XXS, pp:~1400,tp:~65)]]></title><description><![CDATA[<p dir="auto">最近也折腾了一下 Strata. 报告一下配置情况:</p>
<ul>
<li>主机: <a href="https://minisforum.cn/index/product.html?id=25" rel="nofollow ugc">铭凡 um880pro</a>;</li>
<li>内存: 96G DDR5</li>
<li>显卡: 7900xtx, oculink, gen4x4</li>
</ul>
<p dir="auto"><img src="https://upload.lcz.me/uploads/3ba96cd5-5d10-45f0-b6cf-72d6c64a7949.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
<p dir="auto"><img src="https://upload.lcz.me/uploads/d3e69bf5-c287-41b3-b515-8b7adae21377.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
<p dir="auto">配置 (venv 中 rocm 版本 v10.1):</p>
<pre><code> $ cat strata-swift-iq3_xxs.json
{
  "exe": "/tank/github/Strata/engine/strata",
  "args": [
    "--pack",
    "/tank/models/Strata-data/packs/swift-iq3_xxs",
    "--native",
    "/tank/models/Strata-data/models/swift-IQ3_XXS/Swift-Qwen3.8-Flash-Next-GSQ-RCO-IQ3_XXS-00001-of-00002.gguf",
    "--ple-gguf",
    "/tank/models/Strata-data/models/swift-IQ3_XXS/Swift-Qwen3.8-Flash-Next-GSQ-RCO-IQ3_XXS-00001-of-00002.gguf",
    "--expert-profile",
    "/tank/github/Strata/data/expert-profile.bin",
    "--expert-cache",
    "auto",
    "--prefill",
    "32768",
    "--spec",
    "4",
    "--spec-min-p",
    "0.5",
    "--mtp",
    "/tank/models/Strata-data/mtp/rt",
    "--max-context",
    "262144",
    "--kv",
    "int8",
    "--kv-resident",
    "32768"
  ],
  "cwd": "/tank/github/Strata",
  "tokenizer": "/tank/models/Strata-data/packs/swift-iq3_xxs/tokenizer",
  "model_name": "swift-1.5-iq3_xxs",
  "log": "/tank/github/Strata/strata-swift-iq3_xxs.log",
  "lib_dirs": [
    "/tank/github/Strata/.venv/lib/python3.13/site-packages/_rocm_sdk_devel/lib",
    "/tank/github/Strata/.venv/lib/python3.13/site-packages/_rocm_sdk_libraries_gfx110X_dgpu/lib"
  ],
  "port": 8080,
  "backend": "hip",
  "env": {
    "STRATA_HIPBLASLT_TUNING": "/home/bruin/strata-setup/100401-retuned.txt",
    "STRATA_HIPBLASLT_VERBOSE": "1"
  },
  "gpu": 0,
  "gpus_asked": true,
  "host": "0.0.0.0",
  "api_key": "strata",
  "open_browser": false
</code></pre>
<p dir="auto">btw, 双卡的情况下 (peer-device), 速度下降, 应该是我的pcie带宽不够 ,一个gen4x4, 一个gen3x4(雷电3).</p>
]]></description><link>https://lcz.me/topic/2242</link><generator>RSS for Node</generator><lastBuildDate>Mon, 12 Oct 2026 12:24:28 GMT</lastBuildDate><atom:link href="https://lcz.me/topic/2242.rss" rel="self" type="application/rss+xml"/><pubDate>Sat, 10 Oct 2026 06:21:07 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to Strata/单卡7900xtx 配置一例 (IQ3_XXS, pp:~1400,tp:~65) on Sat, 10 Oct 2026 07:35:08 GMT]]></title><description><![CDATA[<p dir="auto">不过长上下文实测有  loop:</p>
<p dir="auto"><img src="https://upload.lcz.me/uploads/4cd06b58-1267-46c9-bc65-22d2c3d118ba.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
]]></description><link>https://lcz.me/post/23737</link><guid isPermaLink="true">https://lcz.me/post/23737</guid><dc:creator><![CDATA[laobenxiong]]></dc:creator><pubDate>Sat, 10 Oct 2026 07:35:08 GMT</pubDate></item></channel></rss>