Pending benchmarksTracking since Oct 20, 2026
Is Sentiment-Analysis-7B state of the art?Pending benchmarks.Announced Oct 20, 2026; no independent results yet.
Microsoft's 7B open-weights model, fine-tuned from Mistral-7B-Instruct-v0.2, that labels a user message about a service, product or agent as Negative, Neutral or Positive and explains why. Trained with reinforcement learning and an LLM judge on about 300 examples without pre-defined labels. MIT license.
Announced facts
The details so far
- Lab
- Microsoft
- Announced
- Oct 20, 2026
- API
- Not yet
- Open weights
- Published
- Parameters
- 7B
- Base model
- mistralai/Mistral-7B-Instruct-v0.2
- Training
- Reinforcement learning with an LLM as judge, on a small internal set of ~300 examples; no pre-defined labels in training
- Output
- First token is the label (Negative, Neutral or Positive), followed by a short explanation
- License
- MIT
- Hugging Face repo created
- Oct 17, 2026 (UTC); announced Oct 20, 2026
Tags marked unsure are not stated on the lab's own pages.
Next steps
What would let us call it
Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.
- Publicly announced with a source
- Usable by the public (API or weights)
- Listed on an independent board we track
- Enough results for a rank in a category
- Holds across two snapshots (crowns only)
Self-reported
What the lab claims
Self-reported. Shown for context; not used in any verdict.
Lab claim: 72.8% accuracy on Microsoft's evaluation, against 62.0% for the Mistral-7B-Instruct-v0.2 base.
Receipts
Sources
REPOSentiment-Analysis-7B on Hugging Facehuggingface.co/microsoft/Sentiment-Analysis-7B · Oct 20, 2026PRESSHuggingPapers post on Xx.com/HuggingPapers/status/2112629159604203995 · Oct 20, 2026Tracking continues until an independent board lists it.