Model comparison
Llama-3.3-70B-Instruct vs Mistral 7B
Llama-3.3-70B-Instruct is the stronger model overall, scoring 30.6 to 23.0 on the Noometry Index.
Last verified . 24 shared benchmarks.
Meta
Mistral AISummary
- They share 24 benchmarks with published results for both. Llama-3.3-70B-Instruct scores higher in 7 categories and Mistral 7B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama-3.3-70B-Instruct leads 30.6 to 7.4.
- The biggest single-benchmark swing is MATH Level 5: 41.6% for Llama-3.3-70B-Instruct and 3.7% for Mistral 7B.
- Llama-3.3-70B-Instruct is cheaper at $0.10 / $0.32 per million input/output tokens, against $0.25 / $0.25 for Mistral 7B.
- Llama-3.3-70B-Instruct accepts more context: 128K tokens versus 8K.
Side by side
| Llama-3.3-70B-Instruct | Mistral 7B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 30.6 | 23.0 |
| Released | 2024-12-06 | 2023-09-27 |
| Weights | Open | Open |
| Context window | 128K | 8K |
| Max output | 4K | 8K |
| Input $ / M tokens | $0.10 | $0.25 |
| Output $ / M tokens | $0.32 | $0.25 |
| Results tracked | 43 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 31.0 (#290), Mistral 7B: 26.4 (#326)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| BigCodeBench Instruct | 46.9% | 19.5% |
| LMArena Coding | 1268 | 1082 |
| BigCodeBench Complete | 57.5% | 27.3% |
| SciCode | 26% | — |
| WeirdML | 14.4% | — |
| LiveBench Coding | 36.6% | — |
| HumanEval+ | — | 36% |
| MBPP+ | — | 42.1% |
Agentic & Tool Use Not comparable
Llama-3.3-70B-Instruct: 25.8 (#105), Mistral 7B: —
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 31.9% | — |
| BALROG | 23% | — |
Reasoning Too close to call
Llama-3.3-70B-Instruct: 14.1 (#327), Mistral 7B: 13.1 (#336)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1067 |
| DTBench | 59.5% | 42.5% |
| Epoch Capabilities Index | 127.33 | 112.21 |
| SimpleBench | 19.9% | — |
| CritPt | 0% | — |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | 50.8% | — |
| LiveBench Data Analysis | 49.5% | — |
| LMCA | 17.5% | — |
| Adversarial NLI | — | 47.1% |
| BIG-Bench Hard | — | 56.1% |
| ForecastBench | 58.6 | — |
| HellaSwag | — | 81% |
| LiveBench | 50.2% | — |
| PIQA | — | 83% |
| WinoGrande | — | 75.3% |
Math Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 15.3 (#298), Mistral 7B: 8.1 (#325)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 5.1% | 0.3% |
| LMArena Math | 1267 | 1085 |
| MATH Level 5 | 41.6% | 3.7% |
| LiveBench Math | 42.2% | — |
| GSM8K | — | 54.4% |
Knowledge Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 30.6 (#226), Mistral 7B: 7.4 (#311)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| GPQA Diamond | 47.4% | 15.2% |
| LMArena Expert | 1225 | 1036 |
| MMLU | 86.3% | 62.5% |
| Confabulations | 22.8% | — |
| Vectara Hallucination Rate | 4.1% | — |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |
Multilingual Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 39.9 (#220), Mistral 7B: 25.8 (#283)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| LMArena Non-English | 1236 | 1012 |
| LMArena Chinese | 1217 | 1009 |
| LMArena French | 1281 | 1037 |
| LMArena German | 1251 | 987 |
| LMArena Japanese | 1150 | 878 |
| LMArena Russian | 1252 | 1018 |
| LMArena Spanish | 1270 | 1026 |
| LMArena Korean | 1143 | — |
Instruction Following Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 71.1 (#157), Mistral 7B: 54.2 (#280)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | 1242 | 1060 |
| LiveBench Instruction Following | 82.7% | — |
Long Context Mistral 7B leads
Llama-3.3-70B-Instruct: 26.4 (#295), Mistral 7B: 32.2 (#271)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| LMArena Longer Query | 1256 | 1060 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Llama-3.3-70B-Instruct leads
Llama-3.3-70B-Instruct: 47.6 (#207), Mistral 7B: 30.7 (#286)
| Benchmark | Llama-3.3-70B-Instruct | Mistral 7B |
|---|---|---|
| LMArena Text | 1274 | 1090 |
| LMArena Creative Writing | 1250 | 1068 |
| LMArena Multi-Turn | 1280 | 1062 |
| LiveBench Language | 39.2% | — |
Frequently asked questions
Is Llama-3.3-70B-Instruct better than Mistral 7B?
Llama-3.3-70B-Instruct is the stronger model overall, scoring 30.6 to 23.0 on the Noometry Index.
Which is cheaper, Llama-3.3-70B-Instruct or Mistral 7B?
Llama-3.3-70B-Instruct is cheaper. It lists at $0.10 per million input tokens and $0.32 per million output tokens; Mistral 7B lists at $0.25 and $0.25.
Is Llama-3.3-70B-Instruct or Mistral 7B better for coding?
Llama-3.3-70B-Instruct scores higher on coding benchmarks: 31.0 versus 26.4 in the Noometry coding category.
Which has the bigger context window?
Llama-3.3-70B-Instruct does, with 128K tokens against 8K.
How many benchmarks do Llama-3.3-70B-Instruct and Mistral 7B share?
24 benchmarks have published results for both models. Llama-3.3-70B-Instruct has 43 scored results on Noometry and Mistral 7B has 37.