sglang-deepseek-v4-flash-sm120

(★ 33)

Reproducible SGLang image for DeepSeek-V4-Flash-0731 on SM120

Owner Repos

sglang-glm53-flash-sm120 ★ 16

SGLang GLM-5.3-Flash on 2x RTX PRO 6000 Blackwell (SM120, TP2): C4 @ 507k-token KV pool, W4A16+FP8-mix quant, HiCache, native MTP — reproducible image, producers, bench harness, receipts

Python
vllm-deepseek-v4-flash-sm120 ★ 9

Reproducible vLLM images for agentic DeepSeek-V4-Flash workloads on 2x RTX PRO 6000 Blackwell (SM120)

Dockerfile

File Explorer

ZIP Download
  • .gitattributes
  • .gitignore
  • AGENTS.md
  • BENCHMARKS.md
  • CHANGELOG.md
  • Containerfile
  • LICENSE
  • README.md
  • release.json
  • RUN.md
  • stack.lock.json

# Use via CDN

jsDelivr

jsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.

Link
Example

Command Glossary

Commands referenced in this DOCs, explained below.

🔍

node

View Details ▼

Server-side JavaScript platform (Node.js).

node {{path/to/file}}

Run a JavaScript file:

node

Start a REPL (interactive shell):

node --watch {{path/to/file}}

Execute the specified file restarting the process when an imported file is changed (requires Node.js version 18.11+):

// repository documentation