Llama 2-70B

Benchmarks, price, context and versions

Direct answer: Llama 2-70B is a model from Meta. 26.3% on GPQA Diamond, per Epoch AI.

✓ Good for

running on your own infrastructure (open weights)

✕ Not the best for

technical reasoning on science topics (#190 of 198 on GPQA Diamond)
calculation and problems with an exact answer (#186 of 186 on OTIS Mock AIME)

This block follows a rule of this section, using source data: if the model is in the top 25% of the metric, the item goes under "Good for"; if it is in the bottom half, it goes under "Not the best for". The middle band appears in neither. Price and context use OpenRouter; scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.

Benchmarks

GPQA Diamond

?what it is

26.3%

highest score on this metric 96.4%

190º / 198 · Epoch AI, latest version we checked

OTIS Mock AIME

highest score on this metric 100%

186º / 186 · Epoch AI, latest version we checked

Key specs

Lab: Meta
Origin: the US
Family: Llama provisional data
Weights?: open
License: no data
Input US$/1M?: no data
Output US$/1M: no data
Context?: no data
Parameters: no data
Input: no data
OpenRouter usage?: no data
Downloads (30 days)?: no data
Use cases provisional data: Text
Dates in sources: 2023, published, per Epoch AI

Version history

2025 PhysicsLM4.2__LlamaCanon-8B-Nemo-1T-lr0.003 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-8B-Nemo-1T-lr0.002 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-3B-Nemo-1T-lr0.003 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-3B-Nemo-1T-lr0.002 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-1B-Nemo-2T-lr0.005 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-1B-Nemo-2T-lr0.003 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-1B-Nemo-1T-lr0.003 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__LlamaCanon-1B-Nemo-1T-lr0.002 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__Llama-8B-Nemo-1T-lr0.003 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__Llama-8B-Nemo-1T-lr0.002 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__Llama-3B-Nemo-1T-lr0.003 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__Llama-3B-Nemo-1T-lr0.002 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__Llama-1B-Nemo-2T-lr0.005 Hugging Face Hub, Hugging Face repo created
2025 PhysicsLM4.2__Llama-1B-Nemo-2T-lr0.003 Hugging Face Hub, Hugging Face repo created

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.

FAQ

What is Llama 2-70B?

Llama 2-70B is a model from Meta, with open weights, used for text.

How much does Llama 2-70B cost?

No data in open sources.

What is the context window of Llama 2-70B?

No data in open sources.

Is Llama 2-70B open source?

Yes, it has open weights. License: no data.

What is Llama 2-70B's GPQA Diamond score?

26.3% on GPQA Diamond, per Epoch AI.

Is Llama 2-70B good for coding?

No data: there is no SWE-bench Verified measurement for it in open sources.

Who made Llama 2-70B?

It was made by Meta, a lab from the US.

provisional data Questions under review: Semrush had no API balance during collection. Answers use source data only.

Models from Meta

See lab

Sources

Epoch AI, CC BY 4.0, latest version we checked.

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.