Llama 2
Llama 2 is an open-weights AI model from Meta with an index score of 95 (GPT-4 = 100), rank 204 of 217.
9579–111Index score#204 of 217 models
–Price per million tokensno public API price found
–Context windowtokens it reads at once
18 Jul 2023ReleasedReplaced by Llama 3
Where it sitson the frontier
GPT-4 reached this score 4 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingno evidence yet
Agentsevidence
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 8 leaderboards
The Model Indexthird-party tests, our method#204of 217 · 95
Epoch Capabilities Indexoverall capability from many benchmarks#184of 231 · 105.6LMArena · Textoverall chat quality, judged by people#314of 362 · 1171LMArena · Instruction followingfollowing detailed instructions#317of 362 · 1135LMArena · Mathsmaths questions in chat#305of 346 · 1137LMArena · Codingcoding questions in chat#321of 357 · 1179LMArena · Creative writingcreative writing#324of 360 · 1111LMArena · Long promptslong, complex requests#319of 340 · 1140Best attop three among current models
Replaced by Llama 3; it no longer leads any category among current models.
Facts and sources
Released18 Jul 2023
Announcementai.meta.com ↗
Release entryLlama 2
Scored asLlama 2-70B
Output speed–
Hugging Facenot linked