Mistral 7B
Mistral 7B is an open-weights AI model from Mistral with an index score of 93 (GPT-4 = 100), provisional.
9377–109Index scoreProvisional · one third-party tester so far; ranked once a second confirms it
–Price per million tokensno public API price found
–Context windowtokens it reads at once
27 Sep 2023ReleasedReplaced by Mistral 7B v0.3
Where it sitson the frontier
GPT-4 reached this score 6 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingno evidence yet
Agentsno evidence yet
Professional workno evidence yet
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 7 leaderboards
Epoch Capabilities Indexoverall capability from many benchmarks#186of 231 · 102.6LMArena · Creative writingcreative writing#335of 360 · 1091LMArena · Instruction followingfollowing detailed instructions#342of 362 · 1087LMArena · Mathsmaths questions in chat#327of 346 · 1083LMArena · Codingcoding questions in chat#338of 357 · 1145LMArena · Textoverall chat quality, judged by people#345of 362 · 1111LMArena · Long promptslong, complex requests#332of 340 · 1100
Best attop three among current models
Replaced by Mistral 7B v0.3; it no longer leads any category among current models.
Facts and sources
Released27 Sep 2023
Announcementmistral.ai ↗
Release entryMistral 7B
Scored asMistral 7B v0.1
Output speed–
Hugging Facenot linked