Llama-3.3-70B-Instruct: benchmark scores, price and ranking
Llama-3.3-70B-Instruct from Meta AI ranks #731 of 743 LLMs on the LLMs Tiger aggregate benchmark score (40.7), based on 11 benchmark results captured up to 2026-10-13.
- Organisation
- Meta AI
- Type
- Open-weight / local model
- Open weights
- Yes
- Released
- 2024-12-06 (source: epoch)
- Cheapest API price
- $0.120 in / $0.300 out per million tokens (via hyperbolic, observed 2026-10-13T19:47Z)
- Aggregate rank
- #731 of 743 (score 40.7)
Relative to other models, Llama-3.3-70B-Instruct is strongest on Aider Code Editing (#37 of 92) and weakest on Fiction.liveBench (#45 of 47).
Across the 11 benchmarks where at least ten models are scored, Llama-3.3-70B-Instruct is in the top 10% on 0 and in the top half on 1.
Its cheapest listed input price, $0.120 per million tokens, is lower than that of 77% of the 427 models with a listed price.
Llama-3.3-70B-Instruct benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Aider Code Editing | 59.4% | #37 of 92 | Aider Aider Code Editing leaderboard | 2026-10-13 |
| BALROG | 23.0% | #19 of 34 | Epoch AI Benchmarking Hub | 2026-10-13 |
| GPQA Diamond | 47.4% | #198 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| LMCA | 17.5% | #78 of 108 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 59.5% | #106 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 5.1% | #157 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
| CritPt | 0.0% | #101 of 115 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SciCode | 26.0% | #96 of 106 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WeirdML | 14.4% | #99 of 109 | Epoch AI Benchmarking Hub | 2026-10-13 |
| SimpleBench | 19.9% | #56 of 61 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Fiction.liveBench | 33.3% | #45 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Llama-3.3-70B-Instruct
Llama-3.3-70B-Instruct is an open-weight model that can run on your own hardware from Meta AI. It was released on 2024-12-06, according to Epoch AI. The cheapest API listing we track charges $0.120 per million input tokens and $0.300 per million output tokens, through hyperbolic. We hold 11 benchmark results for it across 4 areas, and none is in the top tenth of its board.
Where Llama-3.3-70B-Instruct stands
On our aggregate score, Llama-3.3-70B-Instruct is #731 of 743 ranked models, in the lower half. Its score is 40.7; the scale is centred on 50, so it sits 9.3 points below the typical model. That beats 2% of the ranked field. Among the 28 ranked Meta AI models it is #25. The next one up is Llama-3.2-1B-Instruct at 41.1. The next one down is Llama-4-Scout-17B-16E-Instruct at 40.1.
Llama-3.3-70B-Instruct, benchmark by benchmark
Aider Code Editing puts Llama-3.3-70B-Instruct at 59.4%, #37 of 92. That beats 60% of the models tested, but 24.8 points short of Claude 3.5-Sonnet-20241022 (84.2%). The results either side are Mistral Large 2 (2407) (60.2%) and Grok-2 (58.6%). It covers Aider official benchmark. The board median is 55.6%, so Llama-3.3-70B-Instruct is 3.8 points above it. Source: Aider Aider Code Editing leaderboard, captured 2026-10-13.
Llama-3.3-70B-Instruct lands below the middle on BALROG: 23.0%, #19 of 34, with 53% of the field scoring higher. GPT-6 Astra (max) leads at 68.3%; the closest result above is Llama-3.2-90B-Vision-Instruct at 27.3%. It covers long-horizon game-playing agents. The board median is 27.6%, so Llama-3.3-70B-Instruct is 4.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Llama-3.3-70B-Instruct lands below the middle on GPQA Diamond: 47.4%, #198 of 275, with 72% of the field scoring higher. GPT-6 Astra (max) leads at 95.8%; the closest result above is mistral-small-2503 at 47.5%. It covers expert-level science questions. The board median is 71.7%, so Llama-3.3-70B-Instruct is 24.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Llama-3.3-70B-Instruct lands below the middle on LMCA: 17.5%, #78 of 108, with 71% of the field scoring higher. Claude Opus 5.5 (max) leads at 68.2%; the closest result above is gemini-2.5-flash-lite-preview-06-17 (Default thinking length) at 18.1%. It covers language-model capability assessment. The board median is 29.2%, so Llama-3.3-70B-Instruct is 11.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Llama-3.3-70B-Instruct lands below the middle on DTBench: 59.5%, #106 of 146, with 72% of the field scoring higher. Claude Opus 5.5 (max) leads at 98.9%; the closest result above is qwen3-8b at 59.7%. It covers dynamic tool-use benchmark. The board median is 68.8%, so Llama-3.3-70B-Instruct is 9.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
OTIS Mock AIME is a weak spot. Llama-3.3-70B-Instruct scores 5.1%, #157 of 189, behind 83% of the models measured. The leader, Claude Fable 5.1 (max), reaches 100.0%. It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so Llama-3.3-70B-Instruct is 52.6 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
SciCode is a weak spot. Llama-3.3-70B-Instruct scores 26.0%, #96 of 106, behind 90% of the models measured. The leader, Claude Opus 5.5 (max), reaches 66.9%. It covers scientific research code. The board median is 44.3%, so Llama-3.3-70B-Instruct is 18.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
WeirdML is a weak spot. Llama-3.3-70B-Instruct scores 14.4%, #99 of 109, behind 90% of the models measured. The leader, GPT-6 Astra (max), reaches 93.3%. It covers unusual machine-learning tasks. The board median is 41.2%, so Llama-3.3-70B-Instruct is 26.7 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
SimpleBench is a weak spot. Llama-3.3-70B-Instruct scores 19.9%, #56 of 61, behind 90% of the models measured. The leader, Claude Fable 5 (max), reaches 81.9%. It covers trick questions that test common sense. The board median is 41.4%, so Llama-3.3-70B-Instruct is 21.5 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Fiction.liveBench is a weak spot. Llama-3.3-70B-Instruct scores 33.3%, #45 of 47, behind 94% of the models measured. The leader, grok-4-0709, reaches 94.4%. It covers long-context story comprehension (16k). The board median is 61.1%, so Llama-3.3-70B-Instruct is 27.8 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
The full table above also lists the other 1 result, including boards with fewer than ten models, where a rank says little.
Strengths and weaknesses
Software engineering: uneven, ahead of 35% of the field on average across 2 benchmarks. Its best board here is Aider Code Editing (#37 of 92) and its weakest is SciCode (#96 of 106). Reasoning, maths and science: weak, ahead of 16% of the field on average across 4 benchmarks. Its best board here is GPQA Diamond (#198 of 275) and its weakest is SimpleBench (#56 of 61). Agents and tool use: uneven, ahead of 36% of the field on average across 2 benchmarks. Its best board here is BALROG (#19 of 34) and its weakest is DTBench (#106 of 146). General knowledge and chat: weak, ahead of 14% of the field on average across 3 benchmarks. Its best board here is LMCA (#78 of 108) and its weakest is Fiction.liveBench (#45 of 47).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Llama-3.3-70B-Instruct costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on Llama 4 Maverick ($0.188) and claude-3-haiku-20240307 ($0.250).
Release history
The Meta AI model released before it was Llama-3.2-90B-Vision-Instruct on 2024-09-24, which scores 45.9. The next one was Llama 4 Maverick on 2025-04-05, at 42.0.
Reading these numbers
The 11 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (10), Aider Aider Code Editing leaderboard (1). 1 of them come from an official leaderboard and 10 from datasets or third-party runs.
Models ranked nearby
- #728 c4ai-command-r-plus-08-2024 (41.2)
- #729 Llama-3.2-1B-Instruct (41.1)
- #730 open-mistral-7b (41.0)
- #732 claude-2.1 (40.6)
- #733 Llama-4-Scout-17B-16E-Instruct (40.1)
- #734 gemma-3-4b-it (39.9)
More from Meta AI
- Muse Spark 1.3 (max) (#18, 55.1)
- Muse Spark (#29, 53.7)
- Muse Spark 1.1 (xhigh) (#66, 52.4)
- Muse Spark 1.2 (xhigh) (#113, 51.5)
- Llama 3.1-405B (#142, 51.2)
- Muse Wonder (max) (#209, 50.1)
Compare it on the live leaderboard → · How the aggregate score works