Model comparison
Llama 3.2 3B vs Mistral Large
Mistral Large is the stronger model overall, scoring 31.9 to 28.9 on the Noometry Index. Llama 3.2 3B costs 25× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Meta
Mistral AISummary
- They share 17 benchmarks with published results for both. Llama 3.2 3B scores higher in 2 categories and Mistral Large in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Large leads 40.7 to 24.7.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 21.9% for Llama 3.2 3B and 38.4% for Mistral Large.
- Llama 3.2 3B is cheaper at $0.05 / $0.33 per million input/output tokens, against $2 / $6 for Mistral Large.
Side by side
| Llama 3.2 3B | Mistral Large | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 28.9 | 31.9 |
| Released | 2024-09-24 | 2024-02-26 |
| Weights | Open | Open |
| Context window | 131K | 131K |
| Max output | 118K | 16K |
| Input $ / M tokens | $0.05 | $2 |
| Output $ / M tokens | $0.33 | $6 |
| Results tracked | 18 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral Large leads
Llama 3.2 3B: 27.6 (#319), Mistral Large: 34.3 (#240)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| BigCodeBench Instruct | 23.4% | 30% |
| LMArena Coding | 1098 | 1277 |
| BigCodeBench Complete | 28.3% | 38.3% |
| SciCode | — | 36.2% |
| LiveBench Coding | — | 47.1% |
| ALE-Bench | — | 264.7 |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Mistral Large leads
Llama 3.2 3B: 20.1 (#143), Mistral Large: 28.6 (#89)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | 38.4% |
| BALROG | 10.1% | — |
Reasoning Llama 3.2 3B leads
Llama 3.2 3B: 21.0 (#228), Mistral Large: 15.8 (#310)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Hard Prompts | 1095 | 1257 |
| SimpleBench | — | 22.5% |
| CritPt | — | 0% |
| LiveBench Reasoning | — | 43.5% |
| DTBench | — | 65.1% |
| LiveBench Data Analysis | — | 50.1% |
| LMCA | — | 16.7% |
| Epoch Capabilities Index | — | 128.52 |
| ForecastBench | — | 57.1 |
| LiveBench | — | 48.4% |
Math Llama 3.2 3B leads
Llama 3.2 3B: 32.4 (#214), Mistral Large: 18.2 (#291)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Math | 1126 | 1262 |
| OTIS Mock AIME 2024-2025 | — | 8.5% |
| Omni-MATH | — | 28.1% |
| LiveBench Math | — | 42.5% |
| MATH Level 5 | — | 50.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Too close to call
Llama 3.2 3B: 29.7 (#235), Mistral Large: 30.1 (#230)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Expert | 1090 | 1232 |
| GPQA Diamond | — | 51.3% |
| MMLU-Pro | — | 59.9% |
| Confabulations | — | 21.4% |
| Vectara Hallucination Rate | — | 4.5% |
| GPQA (HELM) | — | 43.5% |
| MMLU | — | 80% |
Multilingual Mistral Large leads
Llama 3.2 3B: 26.2 (#281), Mistral Large: 40.0 (#219)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Non-English | 1019 | 1237 |
| LMArena Chinese | 1017 | 1240 |
| LMArena German | 1056 | 1254 |
| LMArena Russian | 949 | 1257 |
| LMArena French | — | 1325 |
| LMArena Japanese | — | 1188 |
| LMArena Korean | — | 1202 |
| LMArena Spanish | — | 1268 |
Instruction Following Mistral Large leads
Llama 3.2 3B: 56.0 (#275), Mistral Large: 67.9 (#191)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Instruction Following | 1089 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
| IFEval | — | 87.7% |
Long Context Mistral Large leads
Llama 3.2 3B: 33.4 (#261), Mistral Large: 38.3 (#199)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1100 | 1261 |
Writing & Preference Mistral Large leads
Llama 3.2 3B: 24.7 (#307), Mistral Large: 40.7 (#242)
| Benchmark | Llama 3.2 3B | Mistral Large |
|---|---|---|
| LMArena Text | 1110 | 1266 |
| LMArena Creative Writing | 1094 | 1243 |
| EQ-Bench Creative Writing | 595 | 985 |
| LMArena Multi-Turn | 1105 | 1260 |
| Short-Story Creative Writing | — | 69% |
| WildBench | — | 80.1% |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is Llama 3.2 3B better than Mistral Large?
Mistral Large is the stronger model overall, scoring 31.9 to 28.9 on the Noometry Index. Llama 3.2 3B costs 25× less per token, which makes it the better buy when Mistral Large's lead doesn't matter for your workload.
Which is cheaper, Llama 3.2 3B or Mistral Large?
Llama 3.2 3B is cheaper. It lists at $0.05 per million input tokens and $0.33 per million output tokens; Mistral Large lists at $2 and $6.
Is Llama 3.2 3B or Mistral Large better for coding?
Mistral Large scores higher on coding benchmarks: 34.3 versus 27.6 in the Noometry coding category.
Which has the bigger context window?
Both accept 131K tokens.
How many benchmarks do Llama 3.2 3B and Mistral Large share?
17 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and Mistral Large has 51.