Meta, open weights
Llama 3.1-70B
Llama 3.1-70B by Meta ranks 308th of 354 ranked models on the Noometry Index as of October 2026, with a score of 29.6. Its strongest category is agentic & tool use, where it ranks 112th. API pricing starts at $0.40 per million input tokens and $0.40 per million output tokens, with a 128K-token context window.
Last verified
Specifications
- Noometry rank
- #308 of 354
- Index score
- 29.6
- Evidence
- Confirmed 35 results
- Provider
Meta- Released
- July 23, 2024
- Weights
- Open weights
- Reasoning
- No
- Context window
- 128K
- Max output
- 4K
- Input price
- $0.40 / M
- Output price
- $0.40 / M
- Blended price
- $0.40 / M
- Output speed
- Not measured
- Value
- #79 of 219
- Knowledge cutoff
- December 2023
- Input
- text
- Hugging Face
- meta-llama/Meta-Llama-3.1-70B-Instruct
Category scores
Each category score combines every public result we have in that category.
- Coding 30.3
- Agentic & Tool Use 25.1
- Reasoning 21.6
- Math 13.5
- Knowledge 24.2
- Multilingual 38.8
- Instruction Following 65.3
- Long Context 37.6
- Writing & Preference 35.4
| Category | Score | Rank | Results |
|---|---|---|---|
| Coding | 30.3 | #296 | 4 |
| Agentic & Tool Use | 25.1 | #112 | 2 |
| Reasoning | 21.6 | #220 | 3 |
| Math | 13.5 | #304 | 4 |
| Knowledge | 24.2 | #269 | 4 |
| Multilingual | 38.8 | #225 | 1 |
| Instruction Following | 65.3 | #223 | 2 |
| Long Context | 37.6 | #214 | 1 |
| Writing & Preference | 35.4 | #267 | 5 |
Strengths and weaknesses
Categories where Llama 3.1-70B places highest and lowest among the models ranked in each, with its score against that category's median.
Strongest categories
| Category | Score | vs median | Rank |
|---|---|---|---|
| Reasoning | 21.6 | −2.0 | #220 of 350, top 63% |
| Long Context | 37.6 | −3.3 | #214 of 296, top 73% |
| Agentic & Tool Use | 25.1 | −5.3 | #112 of 154, top 73% |
Closest competitors
The models ranked just above and below Llama 3.1-70B. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| OLMo 2 Furious 13B | #304 | 29.7 | — | — | Compare |
| Phi 3 Mini 128k Instruct | #305 | 29.7 | — | — | Compare |
| phi-3-medium 14B | #306 | 29.7 | — | — | Compare |
| Gemma 2B | #307 | 29.6 | — | — | Compare |
| Llama 2-13B | #309 | 29.6 | — | — | Compare |
| Claude 3 Opus | #310 | 29.5 | — | — | Compare |
| DBRX | #311 | 29.4 | — | — | Compare |
| Gemma 2 27B | #312 | 29.4 | $0.65 | — | Compare |
Sponsored placements are available on pages like this one. Advertise on Noometry
Benchmark results
Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.
Coding
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| WeirdML | 9% | #115 of 119, top 97% | Epoch AI | ||
| BigCodeBench Instruct | 46.1% | #14 of 64, top 22% | BigCodeBench | 2024-07-23 | |
| LMArena Coding | 1260 | #222 of 294, top 76% | LMArena | 2026-10-08 | |
| BigCodeBench Complete | 54.8% | #20 of 66, top 31% | BigCodeBench | 2024-07-23 |
Agentic & Tool Use
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| TheAgentCompany | 6.9% | #10 of 14, top 72% | Epoch AI | ||
| BALROG | 27.9% | #20 of 35, top 58% | Epoch AI |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Hard Prompts | 1241 | #226 of 297, top 77% | LMArena | 2026-10-08 | |
| DTBench | 60% | #116 of 151, top 77% | Epoch AI | ||
| LMCA | 14.8% | #105 of 125, top 84% | Epoch AI | ||
| Epoch Capabilities Index | 125.92 | #154 of 213, top 73% | Epoch AI | 2024-07-23 |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 3.6% | #154 of 173, top 90% | Epoch AI | 2025-02-25 | |
| Omni-MATH | 21% | #50 of 57, top 88% | HELM Capabilities | ||
| LMArena Math | 1252 | #210 of 285, top 74% | LMArena | 2026-10-08 | |
| MATH Level 5 | 36.7% | #55 of 79, top 70% | Epoch AI | 2025-01-27 |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | 44.2% | #143 of 186, top 77% | Epoch AI | 2025-01-27 | |
| MMLU-Pro | 65.3% | #38 of 58, top 66% | HELM Capabilities | ||
| GPQA (HELM) | 42.6% | #40 of 57, top 71% | HELM Capabilities | ||
| LMArena Expert | 1209 | #218 of 273, top 80% | LMArena | 2026-10-08 | |
| MMLU | 80.1% | #16 of 81, top 20% | Epoch AI |
Multilingual
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Non-English | 1219 | #225 of 297, top 76% | LMArena | 2026-10-08 | |
| LMArena Chinese | 1215 | #225 of 285, top 79% | LMArena | 2026-10-08 | |
| LMArena French | 1261 | #179 of 223, top 81% | LMArena | 2026-10-08 | |
| LMArena German | 1222 | #186 of 231, top 81% | LMArena | 2026-10-08 | |
| LMArena Japanese | 1132 | #179 of 211, top 85% | LMArena | 2026-10-08 | |
| LMArena Korean | 1140 | #182 of 213, top 86% | LMArena | 2026-10-08 | |
| LMArena Russian | 1234 | #221 of 283, top 79% | LMArena | 2026-10-08 | |
| LMArena Spanish | 1253 | #182 of 226, top 81% | LMArena | 2026-10-08 |
Instruction Following
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| IFEval | 82.1% | #33 of 57, top 58% | HELM Capabilities | ||
| LMArena Instruction Following | 1231 | #227 of 298, top 77% | LMArena | 2026-10-08 |
Long Context
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Longer Query | 1241 | #226 of 291, top 78% | LMArena | 2026-10-08 |
Writing & Preference
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Text | 1261 | #222 of 297, top 75% | LMArena | 2026-10-08 | |
| LMArena Creative Writing | 1232 | #220 of 295, top 75% | LMArena | 2026-10-08 | |
| EQ-Bench Creative Writing | 784 | #104 of 115, top 91% | EQ-Bench | ||
| WildBench | 75.8% | #44 of 57, top 78% | HELM Capabilities | ||
| LMArena Multi-Turn | 1256 | #219 of 295, top 75% | LMArena | 2026-10-08 |
API pricing by provider
| Route | Input $/M | Output $/M | Cached input $/M | Checked |
|---|---|---|---|---|
| bedrock | $0.72 | $0.72 | — | 2026-10-10 |
| openrouter | $0.40 | $0.40 | — | 2026-10-10 |
Compare Llama 3.1-70B
- Llama 3.1-70B vs Llama 3-70B
- Llama 3.1-70B vs Gemma 2B
- Llama 3.1-70B vs Llama 2-13B
- Llama 3.1-70B vs phi-3-medium 14B
- Llama 3.1-70B vs Claude 3 Opus
- Llama 3.1-70B vs Phi 3 Mini 128k Instruct
- Llama 3.1-70B vs DBRX
- Llama 3.1-70B vs GPT-6 Astra
- Llama 3.1-70B vs Claude Fable 5.1
- Llama 3.1-70B vs Gemini 3.8 Flash
- Llama 3.1-70B vs Kimi K3
- Llama 3.1-70B vs Grok 4.6
- Llama 3.1-70B vs Qwen3.8 Max
- Llama 3.1-70B vs GLM-5.3
Other Meta models
- Muse Spark 1.354.8
- Muse Spark50.6
- Muse Spark 1.250.3
- Muse Spark 1.149.9
- Muse Glimmer41.7
- Codellama 70b Instruct33.7
- Llama 4 Maverick30.9
- Codellama 34b Instruct30.8
Frequently asked questions
How good is Llama 3.1-70B?
Llama 3.1-70B by Meta ranks 308th of 354 ranked models on the Noometry Index as of October 2026, with a score of 29.6. Its strongest category is agentic & tool use, where it ranks 112th. API pricing starts at $0.40 per million input tokens and $0.40 per million output tokens, with a 128K-token context window.
How much does Llama 3.1-70B cost?
Llama 3.1-70B costs $0.40 per million input tokens and $0.40 per million output tokens on openrouter.
What is Llama 3.1-70B's context window?
Llama 3.1-70B accepts up to 128K tokens of input and can write up to 4K tokens in one response.
Is Llama 3.1-70B open source?
Yes. Llama 3.1-70B's weights are downloadable from Hugging Face (meta-llama/Meta-Llama-3.1-70B-Instruct); check the license for commercial terms.
What are Llama 3.1-70B's strengths and weaknesses?
Relative to other ranked models, Llama 3.1-70B places best in reasoning, long context, agentic & tool use and lowest in math, coding, knowledge.
What is Llama 3.1-70B best at?
Its best category is agentic & tool use, where it ranks 112th on Noometry.