Sage S2.1-Flash open weights
Partly confirmed
The takeaway
On 15 November 2026 the InternLM organisation of the Shanghai AI Laboratory uploaded Sage S2.1-Flash to Hugging Face under the MIT licence, without a press release.
Status
- Claim
Partly confirmed
- Our reporting
- Medium confidence
- Importance
- 2 of 5
- Last verified
- 18 November 2026
Your AI and this story
- GPT-6.2 Astra168 days after its cutoff
- Claude Opus 5.5138 days after its cutoff
- Gemini 4 Hydrogen229 days after its cutoff
- Grok 4.7168 days after its cutoff
None of these four assistants can know about it. The closest, Claude Opus 5.5, stops 138 days before it.
Key facts
- Hugging Face repo internlm/Sage-S2.1-Flash created 2026-11-15 02:34 UTC, last modified Nov 16; model card body empty at the time OrcaRouter checked; architecture tag ‘glm5_next’ (HF API, read Nov 18: 678 downloads, 125 likes)
- Reported size: 320B total / 18B active (OrcaRouter) or 321B (AI Weekly); BF16 weights plus an FP8 variant (AI Weekly)
- Claimed: MLE-bench Lite 90.3, first, vs DeepSeek-V4.1-Flash-1017 89.2 (AI Weekly; results in a separate GitHub repo per OrcaRouter)
What happened
A large Chinese open-weight science model appeared without announcement; found through a web search on 18 Nov, not through the sweep (no lab feed or Techmeme story).
Sources
3 sources from 3 sites. Numbers match the chips in the text.
3 sources: 1 primary, 1 press, 1 reaction
Primary
- Hugging Face: internlm/Sage-S2.1-Flashhuggingface.co, code
Press
- AI Weekly: InternLM’s 321B Sage S2.1 Flash leads MLE-bench Lite at 90.3aiweekly.co, press
Reactions
- OrcaRouter: Sage S2.1 Flash’s quiet releaseorcarouter.ai, discussion
Changes
- Filed. Benchmark claims single-source