Model comparison
Amazon Nova Pro vs Mistral Medium
Mistral Medium is the stronger model overall, scoring 36.3 to 31.0 on the Noometry Index. Amazon Nova Pro costs 2.1× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Mistral AISummary
- They share 22 benchmarks with published results for both. Amazon Nova Pro scores higher in 3 categories and Mistral Medium in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium leads 60.0 to 43.9.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 5.1% for Amazon Nova Pro and 22.7% for Mistral Medium.
- Amazon Nova Pro is cheaper at $0.80 / $3.20 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Amazon Nova Pro accepts more context: 300K tokens versus 262K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| Amazon Nova Pro | Mistral Medium | |
|---|---|---|
| Provider | Amazon | Mistral AI |
| Noometry Index | 31.0 | 36.3 |
| Released | 2024-12-03 | 2023-12-11 |
| Weights | Proprietary | Open |
| Context window | 300K | 262K |
| Max output | 10K | 262K |
| Input $ / M tokens | $0.80 | $1.50 |
| Output $ / M tokens | $3.20 | $7.50 |
| Results tracked | 38 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Amazon Nova Pro: 35.1 (#229), Mistral Medium: 34.2 (#243)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Coding | 1270 | 1434 |
| FrontierCode | — | 8% |
| SciCode | — | 40.2% |
| WeirdML | — | 43.7% |
| LiveBench Coding | 38.1% | — |
| ALE-Bench | — | 763.98 |
Agentic & Tool Use Mistral Medium leads
Amazon Nova Pro: 16.7 (#147), Mistral Medium: 28.3 (#90)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | 37.7% |
| TheAgentCompany | 1.7% | — |
Reasoning Mistral Medium leads
Amazon Nova Pro: 20.0 (#243), Mistral Medium: 24.0 (#167)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1426 |
| Kagi LLM Benchmark | — | 50% |
| CritPt | — | 0% |
| LiveBench Reasoning | 32.6% | — |
| DTBench | — | 75.5% |
| LiveBench Data Analysis | 48.3% | — |
| LMCA | — | 26.1% |
| Surface Evolver Bench | — | 26.9% |
| Epoch Capabilities Index | 123.8 | — |
| LiveBench | 43.5% | — |
Math Too close to call
Amazon Nova Pro: 28.5 (#243), Mistral Medium: 28.1 (#245)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Math | 1252 | 1408 |
| OTIS Mock AIME 2024-2025 | — | 32.2% |
| ProofBench | — | 9% |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |
| MATH Level 5 | — | 81.6% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge Amazon Nova Pro leads
Amazon Nova Pro: 27.4 (#250), Mistral Medium: 25.0 (#265)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| Humanity's Last Exam | 4.4% | 4.5% |
| Vectara Hallucination Rate | 5.1% | 22.7% |
| LMArena Expert | 1211 | 1408 |
| GPQA Diamond | — | 59.5% |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| GPQA (HELM) | 44.6% | — |
| MMLU | 82% | — |
Multimodal Mistral Medium leads
Amazon Nova Pro: 25.0 (#126), Mistral Medium: 35.3 (#88)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Vision | 980 | 1172 |
Multilingual Mistral Medium leads
Amazon Nova Pro: 39.7 (#223), Mistral Medium: 52.1 (#91)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Non-English | 1234 | 1408 |
| LMArena Chinese | 1244 | 1447 |
| LMArena French | 1271 | 1459 |
| LMArena German | 1243 | 1432 |
| LMArena Japanese | 1200 | 1378 |
| LMArena Korean | 1203 | 1380 |
| LMArena Russian | 1240 | 1411 |
| LMArena Spanish | 1182 | 1433 |
Instruction Following Mistral Medium leads
Amazon Nova Pro: 64.9 (#226), Mistral Medium: 73.7 (#116)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Instruction Following | 1235 | 1398 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |
Long Context Mistral Medium leads
Amazon Nova Pro: 38.1 (#205), Mistral Medium: 42.9 (#114)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Longer Query | 1255 | 1406 |
Writing & Preference Mistral Medium leads
Amazon Nova Pro: 43.9 (#226), Mistral Medium: 60.0 (#103)
| Benchmark | Amazon Nova Pro | Mistral Medium |
|---|---|---|
| LMArena Text | 1259 | 1424 |
| LMArena Creative Writing | 1212 | 1391 |
| Short-Story Creative Writing | 60.5% | 77.3% |
| LMArena Multi-Turn | 1246 | 1418 |
| WildBench | 77.7% | — |
| LiveBench Language | 37% | — |
Frequently asked questions
Is Amazon Nova Pro better than Mistral Medium?
Mistral Medium is the stronger model overall, scoring 36.3 to 31.0 on the Noometry Index. Amazon Nova Pro costs 2.1× less per token, which makes it the better buy when Mistral Medium's lead doesn't matter for your workload.
Which is cheaper, Amazon Nova Pro or Mistral Medium?
Amazon Nova Pro is cheaper. It lists at $0.80 per million input tokens and $3.20 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Amazon Nova Pro or Mistral Medium better for coding?
They score almost the same on coding (35.1 vs 34.2); test both on your own repository before choosing.
Which has the bigger context window?
Amazon Nova Pro does, with 300K tokens against 262K.
How many benchmarks do Amazon Nova Pro and Mistral Medium share?
22 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Mistral Medium has 36.