Mistral-7B-Instruct-v0.2
Mistral AIInstruction-tuned 7B chat model from Mistral that follows user prompts in a dialogue format. It was the first widely adopted European open instruct release (Dec 2023) and remains the canonical 7B chat baseline in eval papers and fine-tuning tutorials. Still pulls ~3.2M Hugging Face downloads per month nearly 2.5 years after release.
Mistral 7B Instruct v0.2, 7B dense, 32k context (up from 8k in v0.1); instruct fine-tune of Mistral-7B-v0.2 (base paper arXiv:2310.06825, Oct 2023). >12 months old and superseded by v0.3 and the Ministral/Mistral Small lines, but foundational permissive instruct model; verified live on HF June 2026.
Openness
3 high confidence- weights
- open(Apache-2.0, on HF)
- data
- closed
- code
- closed(no training/post-training pipeline)
- license
- Apache-2.0(OSI permissive)
Apache-2.0 open weights; instruction-tuning data and recipe undisclosed -> open_weights.
- https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2 recorded 2026-06-04
live card; license=apache-2.0; 7B instruct; 32k context; 2,504,301 downloads last month
Adoption
4 high confidence~2.5M downloads last month on HF plus a very large derivative ecosystem (~1,089 finetunes, 1,256 adapters, 349 merges); still a heavily used permissive 7B instruct base years after release.
- https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2 recorded 2026-06-04
2,504,301 downloads last month; 1,089 finetunes / 1,256 adapters
Capability
2 medium confidenceExcellent for a 7B in 2023-24; clearly below 2026 frontier and below modern small instruct models (Phi-4-mini, Qwen 3.x small). Capability low on a frontier-anchored scale.
- https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2 recorded 2026-06-04
7B instruct model card; arXiv:2310.06825
Unchanged since 2026-07-29 (last edited, not re-checked)