MetaLlama 3.1Open SourceArena #133Aug 2, 2024

Meta: Llama 3.1 405B (base)

Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This is the base 405B pre-trained version. It has demonstrated strong performance compared to leading closed-source models in human evaluations. To read more about the model release, click here. Usage of this model is subject to Meta's Acceptable Use Policy.

Context Window
33K
tokens
Max Output
33K
tokens
Released
Aug 2, 2024
Arena Rank
#133
of 305 models
Output Speed
48
tokens/sec
Time to First Token
1.5s
TTFT

Capabilities

👁Vision
🧠Reasoning
🔧Tool Calling
Prompt Caching
🖥Computer Use
🎨Image Generation

Supported Parameters

Frequency Penalty
Reduce repetition
Logit Bias
Adjust token weights
Max Tokens
Output length limit
min_p
Presence Penalty
Encourage new topics
Repetition Penalty
Penalize repeated tokens
Seed
Deterministic outputs
Stop Sequences
Custom stop tokens
Temperature
Controls randomness
Top K
Top-K sampling
Top P
Nucleus sampling

Pricing Comparison

RouterInput / 1MOutput / 1MCached Input / 1M
OpenRouter$4.00$4.00
Vercel AI$0.40$0.40
Martian$4.00$4.00

Benchmarks

Open LLM Leaderboard
AverageOpen LLM Leaderboard
24.02/100
IFEvalOpen LLM Leaderboard
50.58/100
BBHOpen LLM Leaderboard
29.19/100
MATH Lvl 5Open LLM Leaderboard
15.48/100
GPQAOpen LLM Leaderboard
9.51/100
MUSROpen LLM Leaderboard
8.51/100
MMLU-PROOpen LLM Leaderboard
30.86/100
Artificial Analysis
Intelligence IndexArtificial Analysis
51/100
Coding IndexArtificial Analysis
46/100
Math IndexArtificial Analysis
54/100
MMLU-PROArtificial Analysis
0.682/1
GPQA DiamondArtificial Analysis
0.488/1
MATH-500Artificial Analysis
0.738/1
AIME 2024Artificial Analysis
0.097/1
LiveCodeBenchArtificial Analysis
0.398/1
SciCodeArtificial Analysis
0.162/1
Aider Code Editing
Aider EditingAider Code Editing
26.3/100

Model IDs

OpenRoutermeta-llama/llama-3.1-405b
Compare with another model

Related Models