Model comparison
Mistral 7B vs Sonar
Sonar is the stronger model overall, scoring 38.5 to 23.0 on the Noometry Index. Mistral 7B costs 4.0× less per token, which makes it the better buy when Sonar's lead doesn't matter for your workload.
Last verified . 0 shared benchmarks.
Mistral AI
PerplexitySummary
- The widest gap is in math, where Sonar leads 33.7 to 8.1.
- Mistral 7B is cheaper at $0.25 / $0.25 per million input/output tokens, against $1 / $1 for Sonar.
- Sonar accepts more context: 128K tokens versus 8K.
- Mistral 7B has downloadable open weights; the other is API-only.
Side by side
| Mistral 7B | Sonar | |
|---|---|---|
| Provider | Mistral AI | Perplexity |
| Noometry Index | 23.0 | 38.5 |
| Released | 2023-09-27 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | 8K | 128K |
| Max output | 8K | 4K |
| Input $ / M tokens | $0.25 | $1 |
| Output $ / M tokens | $0.25 | $1 |
| Results tracked | 37 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Sonar leads
Mistral 7B: 26.4 (#326), Sonar: 35.7 (#221)
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| BigCodeBench Instruct | 19.5% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1082 | — |
| BigCodeBench Complete | 27.3% | — |
| HumanEval+ | 36% | — |
| MBPP+ | 42.1% | — |
Reasoning Sonar leads
Mistral 7B: 13.1 (#336), Sonar: 21.1 (#227)
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1067 | — |
| DTBench | 42.5% | — |
| LiveBench Data Analysis | — | 37.9% |
| Adversarial NLI | 47.1% | — |
| BIG-Bench Hard | 56.1% | — |
| Epoch Capabilities Index | 112.21 | — |
| HellaSwag | 81% | — |
| LiveBench | — | 46.9% |
| PIQA | 83% | — |
| WinoGrande | 75.3% | — |
Math Sonar leads
Mistral 7B: 8.1 (#325), Sonar: 33.7 (#200)
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.3% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1085 | — |
| MATH Level 5 | 3.7% | — |
| GSM8K | 54.4% | — |
Knowledge Not comparable
Mistral 7B: 7.4 (#311), Sonar: —
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| GPQA Diamond | 15.2% | — |
| LMArena Expert | 1036 | — |
| ARC (AI2) Challenge | 78.6% | — |
| BoolQ | 87.4% | — |
| MMLU | 62.5% | — |
| OpenBookQA | 79.8% | — |
| TriviaQA | 75.2% | — |
Multilingual Not comparable
Mistral 7B: 25.8 (#283), Sonar: —
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| LMArena Non-English | 1012 | — |
| LMArena Chinese | 1009 | — |
| LMArena French | 1037 | — |
| LMArena German | 987 | — |
| LMArena Japanese | 878 | — |
| LMArena Russian | 1018 | — |
| LMArena Spanish | 1026 | — |
Instruction Following Sonar leads
Mistral 7B: 54.2 (#280), Sonar: 71.4 (#150)
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| LMArena Instruction Following | 1060 | — |
Long Context Not comparable
Mistral 7B: 32.2 (#271), Sonar: —
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| LMArena Longer Query | 1060 | — |
Writing & Preference Sonar leads
Mistral 7B: 30.7 (#286), Sonar: 52.6 (#167)
| Benchmark | Mistral 7B | Sonar |
|---|---|---|
| LMArena Text | 1090 | — |
| LMArena Creative Writing | 1068 | — |
| LMArena Multi-Turn | 1062 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Mistral 7B better than Sonar?
Sonar is the stronger model overall, scoring 38.5 to 23.0 on the Noometry Index. Mistral 7B costs 4.0× less per token, which makes it the better buy when Sonar's lead doesn't matter for your workload.
Which is cheaper, Mistral 7B or Sonar?
Mistral 7B is cheaper. It lists at $0.25 per million input tokens and $0.25 per million output tokens; Sonar lists at $1 and $1.
Is Mistral 7B or Sonar better for coding?
Sonar scores higher on coding benchmarks: 35.7 versus 26.4 in the Noometry coding category.
Which has the bigger context window?
Sonar does, with 128K tokens against 8K.
How many benchmarks do Mistral 7B and Sonar share?
0 benchmarks have published results for both models. Mistral 7B has 37 scored results on Noometry and Sonar has 7.