MA
Meta AI
Language model · United States ·Jul 2024

Llama 3.1-405B

Language Open (Restricted)
405B
Parameters
24
Benchmarks

About Llama 3.1-405B

Llama 3.1-405B is an AI model developed by Meta AI, in the language category, released in 2024, made available as an open-weights model with 405B params.

On this page you'll find Llama 3.1-405B's full specifications, including 24 benchmark results. Review provider pricing and benchmark scores below, or compare Llama 3.1-405B head-to-head with other language models.

Links

Run Llama 3.1-405B locally — what it takes

Llama 3.1-405B (405B) is too large to run on a single consumer or Apple machine. At Q4 its weights alone are about 227 GB — it needs a multi-GPU / datacenter setup:

  • ≈ 1× AMD Instinct MI325X
  • ≈ 2× NVIDIA B200

Lower quantization (Q3/Q2) or CPU + system-RAM offload (e.g. ktransformers) reduces the requirement at the cost of speed.

Software support

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Community

Would you run Llama 3.1-405B again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

Llama 3.1-405B benchmark scores
Benchmark Category Score Variant
HumanEval code 95.9% instruct-pass@10-cited-tulu3
HumanEval code 89% instruct-cited-nvlm
HumanEval code 89% instruct-cited-minimax-text
HumanEval+ code 90.3% instruct-pass@10-cited-tulu3
MBPP+ code 73% instruct-plus-cited-minimax-text
AlpacaEval 2 general 38.5% instruct-lc-win-cited-tulu3
ArenaHard general 63.5% instruct-cited-minimax-text
IFEval general 88.4% instruct-cited-tulu3
IFEval general 86.4% instruct-avg-cited-minimax-text
MMLU general 88.6% instruct-cited-minimax-text
MMLU general 88% instruct-cited-tulu3
MMLU general 87.3% instruct-cited-nvlm
MMLU-Pro general 73.3% instruct-cited-minimax-text
GSM8K math 96.8% instruct-cited-nvlm
GSM8K math 96.7% instruct-cited-minimax-text
GSM8K math 95.4% instruct-cited-tulu3
MATH math 73.8% instruct-cited-nvlm
MATH math 73.8% instruct-cited-minimax-text
MATH math 66.6% instruct-cited-tulu3
BIG-Bench Hard reasoning 87.1% instruct-cited-tulu3
DROP reasoning 92.5% instruct-f1-cited-minimax-text
GPQA Diamond reasoning 50.7% instruct-diamond-cited-minimax-text
C-SimpleQA safety 54.7% instruct-cited-minimax-text
SimpleQA safety 23.2% instruct-cited-minimax-text

More models like Llama 3.1-405B

Frequently asked questions

What is Llama 3.1-405B?

Llama 3.1-405B is an AI model developed by Meta AI, in the language category, released in 2024. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created Llama 3.1-405B?

Llama 3.1-405B was developed by Meta AI and released in 2024.

Is Llama 3.1-405B open source or proprietary?

Llama 3.1-405B ships as an open-weights model: you can download and self-host the weights, though the licence may place some restrictions on use.

How much does Llama 3.1-405B cost?

Pricing for Llama 3.1-405B depends on the provider. See the providers table on this page for the latest API rates.

How does Llama 3.1-405B perform on benchmarks?

Llama 3.1-405B is benchmarked across 24 evaluations on GenAIList, including HumanEval (95.9%). See the full benchmark table below and compare it with other models.

No videos yet. Know a good one? Add it below.

Know a great video about Llama 3.1-405B?