Meta, open weights
Llama 2-70B
Llama 2-70B by Meta ranks 349th of 354 ranked models on the Noometry Index as of October 2026, with a score of 24.4. Its strongest category is long context, where it ranks 270th.
Last verified
Specifications
- Noometry rank
- #349 of 354
- Index score
- 24.4
- Evidence
- Confirmed 35 results
- Provider
Meta- Released
- July 18, 2023
- Weights
- Open weights
- Reasoning
- Unknown
- Context window
- —
- Max output
- —
- Input price
- Not listed
- Output price
- Not listed
- Blended price
- Not listed
- Output speed
- Not measured
- Value
- Not ranked
- Knowledge cutoff
- Unknown
Category scores
Each category score combines every public result we have in that category.
- Coding 31.4
- Reasoning 14.4
- Math 8.1
- Knowledge 7.4
- Multilingual 27.7
- Instruction Following 54.9
- Long Context 32.3
- Writing & Preference 32.3
| Category | Score | Rank | Results |
|---|---|---|---|
| Coding | 31.4 | #286 | 1 |
| Reasoning | 14.4 | #325 | 2 |
| Math | 8.1 | #326 | 3 |
| Knowledge | 7.4 | #310 | 2 |
| Multilingual | 27.7 | #274 | 1 |
| Instruction Following | 54.9 | #278 | 1 |
| Long Context | 32.3 | #270 | 1 |
| Writing & Preference | 32.3 | #279 | 3 |
Strengths and weaknesses
Categories where Llama 2-70B 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 |
|---|---|---|---|
| Coding | 31.4 | −7.3 | #286 of 340, top 85% |
| Writing & Preference | 32.3 | −21.5 | #279 of 312, top 90% |
| Instruction Following | 54.9 | −16.4 | #278 of 305, top 92% |
Closest competitors
The models ranked just above and below Llama 2-70B. When scores are this close, price and speed are often the better way to choose.
| Model | Rank | Score | Blended $/M | Speed | |
|---|---|---|---|---|---|
| Claude 2.1 | #345 | 25.2 | — | — | Compare |
| Claude 2 | #346 | 25.0 | — | — | Compare |
| DeepSeek LLM 67B | #347 | 24.9 | — | — | Compare |
| Llama 13b | #348 | 24.4 | — | — | Compare |
| GPT-3.5-turbo | #350 | 23.2 | $0.75 | — | Compare |
| Mistral 7B | #351 | 23.0 | $0.25 | — | Compare |
| Llama 3.1-8B | #352 | 23.0 | $0.0575 | — | Compare |
| Gemma 3 1B | #353 | 21.1 | — | — | 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 |
|---|---|---|---|---|---|
| LMArena Coding | 1079 | #278 of 294, top 95% | LMArena | 2026-10-08 |
Reasoning
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Hard Prompts | 1073 | #274 of 297, top 93% | LMArena | 2026-10-08 | |
| DTBench | 41.6% | #151 of 151, top 100% | Epoch AI | ||
| BIG-Bench Hard | 58.5% | Epoch AI | |||
| BIG-Bench Hard | 64.9% | #11 of 27, top 41% | Epoch AI | ||
| CommonsenseQA 2.0 | 50% | #2 of 2 | Epoch AI | ||
| Epoch Capabilities Index | 113.79 | #185 of 213, top 87% | Epoch AI | 2023-07-18 | |
| ForecastBench | 51.4 | #71 of 72, top 99% | Epoch AI | ||
| HellaSwag | 85.3% | #8 of 29, top 28% | Epoch AI | ||
| LAMBADA | 78.9% | #2 of 9, top 23% | Epoch AI | ||
| PIQA | 82.8% | #13 of 27, top 49% | Epoch AI | ||
| WinoGrande | 80.2% | #15 of 43, top 35% | Epoch AI |
Math
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | #173 of 173, top 100% | Epoch AI | 2025-02-25 | |
| LMArena Math | 1091 | #267 of 285, top 94% | LMArena | 2026-10-08 | |
| MATH Level 5 | 3.3% | #79 of 79, top 100% | Epoch AI | 2025-01-27 | |
| GSM8K | 69.6% | #15 of 38, top 40% | Epoch AI | ||
| GSM8K | 58.7% | Epoch AI |
Knowledge
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | 26.3% | #178 of 186, top 96% | Epoch AI | 2025-01-27 | |
| LMArena Expert | 1039 | #265 of 273, top 98% | LMArena | 2026-10-08 | |
| ARC (AI2) Challenge | 78.3% | #15 of 39, top 39% | Epoch AI | ||
| BoolQ | 88.6% | #3 of 23, top 14% | Epoch AI | ||
| MMLU | 69.9% | #46 of 81, top 57% | Epoch AI | ||
| MMLU | 59.9% | Epoch AI | |||
| OpenBookQA | 60.2% | #11 of 19, top 58% | Epoch AI | ||
| TriviaQA | 87.6% | Best of 25 | Epoch AI |
Multilingual
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Non-English | 1045 | #274 of 297, top 93% | LMArena | 2026-10-08 | |
| LMArena Chinese | 995 | #279 of 285, top 98% | LMArena | 2026-10-08 | |
| LMArena French | 1090 | #215 of 223, top 97% | LMArena | 2026-10-08 | |
| LMArena German | 1041 | #223 of 231, top 97% | LMArena | 2026-10-08 | |
| LMArena Japanese | 927 | #208 of 211, top 99% | LMArena | 2026-10-08 | |
| LMArena Korean | 964 | #205 of 213, top 97% | LMArena | 2026-10-08 | |
| LMArena Russian | 1083 | #261 of 283, top 93% | LMArena | 2026-10-08 | |
| LMArena Spanish | 1143 | #207 of 226, top 92% | LMArena | 2026-10-08 |
Instruction Following
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Instruction Following | 1071 | #274 of 298, top 92% | LMArena | 2026-10-08 |
Long Context
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Longer Query | 1062 | #275 of 291, top 95% | LMArena | 2026-10-08 |
Writing & Preference
| Benchmark | Score | Position | Setting | Source | Date |
|---|---|---|---|---|---|
| LMArena Text | 1115 | #269 of 297, top 91% | LMArena | 2026-10-08 | |
| LMArena Creative Writing | 1075 | #273 of 295, top 93% | LMArena | 2026-10-08 | |
| LMArena Multi-Turn | 1088 | #268 of 295, top 91% | LMArena | 2026-10-08 |
Compare Llama 2-70B
- Llama 2-70B vs Llama 13b
- Llama 2-70B vs GPT-3.5-turbo
- Llama 2-70B vs DeepSeek LLM 67B
- Llama 2-70B vs Mistral 7B
- Llama 2-70B vs Claude 2
- Llama 2-70B vs Llama 3.1-8B
- Llama 2-70B vs GPT-6 Astra
- Llama 2-70B vs Claude Fable 5.1
- Llama 2-70B vs Gemini 3.8 Flash
- Llama 2-70B vs Kimi K3
- Llama 2-70B vs Grok 4.6
- Llama 2-70B vs Qwen3.8 Max
- Llama 2-70B vs GLM-5.3
- Llama 2-70B vs DeepSeek V4 Pro
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 2-70B?
Llama 2-70B by Meta ranks 349th of 354 ranked models on the Noometry Index as of October 2026, with a score of 24.4. Its strongest category is long context, where it ranks 270th.
Is Llama 2-70B open source?
Yes. Llama 2-70B's weights are downloadable; check the license for commercial terms.
What are Llama 2-70B's strengths and weaknesses?
Relative to other ranked models, Llama 2-70B places best in coding, writing & preference, instruction following and lowest in math, knowledge, reasoning.
What is Llama 2-70B best at?
Its best category is long context, where it ranks 270th on Noometry.