Model comparison
Llama 3.1-70B vs Mistral 7B
Llama 3.1-70B is the stronger model overall, scoring 29.4 to 22.9 on the Noometry Index. Mistral 7B costs 1.6× less per token, which makes it the better buy when Llama 3.1-70B's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Meta
Mistral AISummary
- They share 24 benchmarks with published results for both. Llama 3.1-70B scores higher in 8 categories and Mistral 7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 3.1-70B leads 24.1 to 7.3.
- The biggest single-benchmark swing is MATH Level 5: 36.7% for Llama 3.1-70B and 3.7% for Mistral 7B.
- Mistral 7B is cheaper at $0.25 / $0.25 per million input/output tokens, against $0.40 / $0.40 for Llama 3.1-70B.
- Llama 3.1-70B accepts more context: 128K tokens versus 8K.
Side by side
| Llama 3.1-70B | Mistral 7B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 29.4 | 22.9 |
| Released | 2024-07-23 | 2023-09-27 |
| Weights | Open | Open |
| Context window | 128K | 8K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.40 | $0.25 |
| Output $ / M tokens | $0.40 | $0.25 |
| Results tracked | 35 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3.1-70B leads
Llama 3.1-70B: 29.8 (#292), Mistral 7B: 25.9 (#321)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| BigCodeBench Instruct | 46.1% | 19.5% |
| LMArena Coding | 1260 | 1082 |
| BigCodeBench Complete | 54.8% | 27.3% |
| WeirdML | 9% | — |
| HumanEval+ | — | 36% |
| MBPP+ | — | 42.1% |
Agentic & Tool Use Not comparable
Llama 3.1-70B: 24.9 (#104), Mistral 7B: —
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| TheAgentCompany | 6.9% | — |
| BALROG | 27.9% | — |
Reasoning Llama 3.1-70B leads
Llama 3.1-70B: 20.8 (#215), Mistral 7B: 12.6 (#331)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1067 |
| DTBench | 60% | 42.5% |
| Epoch Capabilities Index | 125.91 | 111.97 |
| Chess Puzzles | — | 0% |
| LMCA | 14.8% | — |
| Adversarial NLI | — | 47.1% |
| BIG-Bench Hard | — | 56.1% |
| HellaSwag | — | 81% |
| PIQA | — | 83% |
| WinoGrande | — | 75.3% |
Math Llama 3.1-70B leads
Llama 3.1-70B: 12.7 (#298), Mistral 7B: 7.6 (#319)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 3.6% | 0.3% |
| LMArena Math | 1252 | 1085 |
| MATH Level 5 | 36.7% | 3.7% |
| Omni-MATH | 21% | — |
| GSM8K | — | 54.4% |
Knowledge Llama 3.1-70B leads
Llama 3.1-70B: 24.1 (#264), Mistral 7B: 7.3 (#306)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| GPQA Diamond | 44.2% | 15.2% |
| LMArena Expert | 1209 | 1036 |
| MMLU | 80.1% | 62.5% |
| MMLU-Pro | 65.3% | — |
| GPQA (HELM) | 42.6% | — |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |
Multilingual Llama 3.1-70B leads
Llama 3.1-70B: 38.9 (#220), Mistral 7B: 26.0 (#278)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1219 | 1012 |
| LMArena Chinese | 1215 | 1009 |
| LMArena French | 1260 | 1036 |
| LMArena German | 1222 | 987 |
| LMArena Japanese | 1132 | 878 |
| LMArena Russian | 1234 | 1018 |
| LMArena Spanish | 1253 | 1026 |
| LMArena Korean | 1140 | — |
Instruction Following Llama 3.1-70B leads
Llama 3.1-70B: 65.5 (#218), Mistral 7B: 54.5 (#275)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1231 | 1060 |
| IFEval | 82.1% | — |
Long Context Llama 3.1-70B leads
Llama 3.1-70B: 37.6 (#208), Mistral 7B: 32.2 (#266)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1241 | 1060 |
Writing & Preference Llama 3.1-70B leads
Llama 3.1-70B: 35.6 (#262), Mistral 7B: 30.9 (#281)
| Benchmark | Llama 3.1-70B | Mistral 7B |
|---|---|---|
| LMArena Text | 1261 | 1090 |
| LMArena Creative Writing | 1232 | 1068 |
| LMArena Multi-Turn | 1256 | 1062 |
| EQ-Bench Creative Writing | 784 | — |
| WildBench | 75.8% | — |
Frequently asked questions
Is Llama 3.1-70B better than Mistral 7B?
Llama 3.1-70B is the stronger model overall, scoring 29.4 to 22.9 on the Noometry Index. Mistral 7B costs 1.6× less per token, which makes it the better buy when Llama 3.1-70B's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-70B or Mistral 7B?
Mistral 7B is cheaper. It lists at $0.25 per million input tokens and $0.25 per million output tokens; Llama 3.1-70B lists at $0.40 and $0.40.
Is Llama 3.1-70B or Mistral 7B better for coding?
Llama 3.1-70B scores higher on coding benchmarks: 29.8 versus 25.9 in the Noometry coding category.
Which has the bigger context window?
Llama 3.1-70B does, with 128K tokens against 8K.
How many benchmarks do Llama 3.1-70B and Mistral 7B share?
24 benchmarks have published results for both models. Llama 3.1-70B has 35 scored results on Noometry and Mistral 7B has 37.