Model comparison
Mistral Large vs Sonar
Sonar is the stronger model overall, scoring 38.5 to 31.9 on the Noometry Index.
Last verified . 7 shared benchmarks.
Mistral AI
PerplexitySummary
- They share 7 benchmarks with published results for both. Mistral Large scores higher in 0 categories and Sonar in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Sonar leads 33.7 to 18.2.
- The biggest single-benchmark swing is LiveBench Data Analysis: 50.1% for Mistral Large and 37.9% for Sonar.
- Sonar is cheaper at $1 / $1 per million input/output tokens, against $2 / $6 for Mistral Large.
- Mistral Large accepts more context: 131K tokens versus 128K.
- Mistral Large has downloadable open weights; the other is API-only.
Side by side
| Mistral Large | Sonar | |
|---|---|---|
| Provider | Mistral AI | Perplexity |
| Noometry Index | 31.9 | 38.5 |
| Released | 2024-02-26 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | 131K | 128K |
| Max output | 16K | 4K |
| Input $ / M tokens | $2 | $1 |
| Output $ / M tokens | $6 | $1 |
| Results tracked | 51 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
Mistral Large: 34.3 (#240), Sonar: 35.7 (#221)
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LiveBench Coding | 47.1% | 35.1% |
| SciCode | 36.2% | — |
| BigCodeBench Instruct | 30% | — |
| LMArena Coding | 1277 | — |
| BigCodeBench Complete | 38.3% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |
Agentic & Tool Use Not comparable
Mistral Large: 28.6 (#89), Sonar: —
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
Reasoning Sonar leads
Mistral Large: 15.8 (#310), Sonar: 21.1 (#227)
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LiveBench Reasoning | 43.5% | 46.3% |
| LiveBench Data Analysis | 50.1% | 37.9% |
| LiveBench | 48.4% | 46.9% |
| SimpleBench | 22.5% | — |
| CritPt | 0% | — |
| LMArena Hard Prompts | 1257 | — |
| DTBench | 65.1% | — |
| LMCA | 16.7% | — |
| Epoch Capabilities Index | 128.52 | — |
| ForecastBench | 57.1 | — |
Math Sonar leads
Mistral Large: 18.2 (#291), Sonar: 33.7 (#200)
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LiveBench Math | 42.5% | 41.6% |
| OTIS Mock AIME 2024-2025 | 8.5% | — |
| Omni-MATH | 28.1% | — |
| LMArena Math | 1262 | — |
| MATH Level 5 | 50.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Not comparable
Mistral Large: 30.1 (#230), Sonar: —
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| GPQA Diamond | 51.3% | — |
| MMLU-Pro | 59.9% | — |
| Confabulations | 21.4% | — |
| Vectara Hallucination Rate | 4.5% | — |
| GPQA (HELM) | 43.5% | — |
| LMArena Expert | 1232 | — |
| MMLU | 80% | — |
Multilingual Not comparable
Mistral Large: 40.0 (#219), Sonar: —
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LMArena Non-English | 1237 | — |
| LMArena Chinese | 1240 | — |
| LMArena French | 1325 | — |
| LMArena German | 1254 | — |
| LMArena Japanese | 1188 | — |
| LMArena Korean | 1202 | — |
| LMArena Russian | 1257 | — |
| LMArena Spanish | 1268 | — |
Instruction Following Sonar leads
Mistral Large: 67.9 (#191), Sonar: 71.4 (#150)
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LiveBench Instruction Following | 67.9% | 76.2% |
| IFEval | 87.7% | — |
| LMArena Instruction Following | 1249 | — |
Long Context Not comparable
Mistral Large: 38.3 (#199), Sonar: —
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LMArena Longer Query | 1261 | — |
Writing & Preference Sonar leads
Mistral Large: 40.7 (#242), Sonar: 52.6 (#167)
| Benchmark | Mistral Large | Sonar |
|---|---|---|
| LiveBench Language | 39.4% | 44.1% |
| LMArena Text | 1266 | — |
| LMArena Creative Writing | 1243 | — |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| WildBench | 80.1% | — |
| LMArena Multi-Turn | 1260 | — |
Frequently asked questions
Is Mistral Large better than Sonar?
Sonar is the stronger model overall, scoring 38.5 to 31.9 on the Noometry Index.
Which is cheaper, Mistral Large or Sonar?
Sonar is cheaper. It lists at $1 per million input tokens and $1 per million output tokens; Mistral Large lists at $2 and $6.
Is Mistral Large or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.7 versus 34.3 in the Noometry coding category.
Which has the bigger context window?
Mistral Large does, with 131K tokens against 128K.
How many benchmarks do Mistral Large and Sonar share?
7 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Sonar has 7.