Llama 3-8B
Llama 3-8B is an open-weights AI model from Meta with an index score of 89 (GPT-4 = 100), rank 212 of 217.
8971–108Index score#212 of 217 models
–Price per million tokensno public API price found
–Context windowtokens it reads at once
18 Apr 2024ReleasedReplaced by Llama 3.1-8B
Where it sitson the frontier
GPT-4 reached this score 13 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#212of 217 · 89
LMArena · Creative writingcreative writing#276of 360 · 1196Epoch Capabilities Indexoverall capability from many benchmarks#181of 231 · 109LMArena · Textoverall chat quality, judged by people#285of 362 · 1224LMArena · Instruction followingfollowing detailed instructions#289of 362 · 1193LMArena · Codingcoding questions in chat#286of 357 · 1253LMArena · Mathsmaths questions in chat#283of 346 · 1193LMArena · Long promptslong, complex requests#290of 340 · 1210Best attop three among current models
Replaced by Llama 3.1-8B; it no longer leads any category among current models.
Facts and sources
Released18 Apr 2024
Announcementnot linked yet
Release entry–
Scored asLlama 3-8B
Output speed–
Hugging Facenot linked