Skip to content

Instantly share code, notes, and snippets.

@jazzqi

jazzqi/hermes-free-models-report.md

Created April 22, 2026 16:29
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save jazzqi/4de61d9602c135702637289d7dc332a3 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save jazzqi/4de61d9602c135702637289d7dc332a3 to your computer and use it in GitHub Desktop.
Download ZIP
Hermes Free Model Support Report
Raw
hermes-free-models-report.md

Hermes Free Model Support Report

Supported Free Models

Model Provider Context Size Output Size Reasoning Speed (tokens/s) Notes
glm-5.1 nvidia 200K 131K ✅ ~100+ Latest GLM, supports tool calls
DeepSeek-R1 nvidia 128K 128K ✅ ~80 Strong reasoning
llama-3.1-nemotron-70b nvidia 128K 64K ❌ ~100 Optimized for chat
chatqa-2 nvidia 128K 64K ❌ ~60 Question answering
mimo-v2-pro nous 128K 64K ❌ ~50 Code-focused, slower
DeepSeek-V3.2-Thinking nvidia 128K 128K ✅ ~70 Balanced reasoning
qwen-flash 302ai 1M 32K ❌ ~40 Very large context

Recommendations

  • General Use: nvidia/llama-3.1-nemotron-70b (BEST PERFORMANCE)
  • Code/Technical: xiaomi/mimo-v2-pro (DEEP CODE UNDERSTANDING)
  • Complex Reasoning: deepseek-ai/DeepSeek-R1 (POWERFUL LOGIC)
  • Long Context: qwen-flash (1M CONTEXT WINDOW)

Last updated: 2026-04-22

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment