Llama-3.3-70B-Instruct

Language Meta Partly verified
  • In Input modalities: Text Out Output modalities: Text
  • Reasoning
  • Tool use
  • Structured output
  • Open weights

Prices

0 offerings · prices as listed

No priced offering is listed for this model yet.

Details

Knowledge cutoff
Dec 2023
Family
llama

Not stated on Meta’s pages: max input, reasoning efforts, reasoning can be off, max thinking budget, built-in tools, licence, aliases.

  • Parameters 70.6B
  • Dense
  • Native precision BF16
  • API context 125K (maker’s API)
My hardware GeForce RTX 4090 24 GB · 1008 GB/s · 64 GB DDR5 dual channel · 89.6 GB/s

Memory isn’t estimated for this type: it depends on resolution, length and pipeline. Its weights:

PrecisionWeights
BF16 (native)maker’s file131.4 GB

Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated

Cited scores

From third-party leaderboards, not measured by us
  • LMArena Text 1,318

    #143 of 163 listed models · CI 1,315–1,322 · 54,493 votes

  • Epoch AI · GPQA Diamond 47.4%

    #135 of 168 listed models · ±2.85

  • Epoch AI · MATH Level 5 41.6%

    #46 of 64 listed models · ±1.18

  • Epoch AI · OTIS Mock AIME 2024–2025 5.1%

    #145 of 159 listed models · ±2.36

  • LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 28, 2026
  • Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 18 – Nov 4, 2026
  • changes we made

No changes recorded yet.

Sources

Meta’s official policies: data use, privacy, terms →

Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data

Meta · United States · ai.meta.com

Snapshots, the same family and other models from this provider