Model comparison
Mistral Medium vs Nova 2 Lite
Nova 2 Lite is the stronger model overall, scoring 39.7 to 36.3 on the Noometry Index.
Last verified . 19 shared benchmarks.
Mistral AISummary
- They share 19 benchmarks with published results for both. Mistral Medium scores higher in 5 categories and Nova 2 Lite in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Nova 2 Lite leads 43.0 to 25.0.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 22.7% for Mistral Medium and 5.1% for Nova 2 Lite.
- Nova 2 Lite is cheaper at $0.30 / $2.50 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium.
- Nova 2 Lite accepts more context: 1M tokens versus 262K.
- Mistral Medium has downloadable open weights; the other is API-only.
Side by side
| Mistral Medium | Nova 2 Lite | |
|---|---|---|
| Provider | Mistral AI | Amazon |
| Noometry Index | 36.3 | 39.7 |
| Released | 2023-12-11 | 2025-12-01 |
| Weights | Open | Proprietary |
| Context window | 262K | 1M |
| Max output | 262K | 64K |
| Input $ / M tokens | $1.50 | $0.30 |
| Output $ / M tokens | $7.50 | $2.50 |
| Results tracked | 36 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nova 2 Lite leads
Mistral Medium: 34.2 (#243), Nova 2 Lite: 40.7 (#134)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Coding | 1434 | 1385 |
| FrontierCode | 8% | — |
| SciCode | 40.2% | — |
| WeirdML | 43.7% | — |
| ALE-Bench | 763.98 | — |
Agentic & Tool Use Mistral Medium leads
Mistral Medium: 28.3 (#90), Nova 2 Lite: 24.1 (#121)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| Berkeley Function Calling Leaderboard | 37.7% | 27.1% |
Reasoning Nova 2 Lite leads
Mistral Medium: 24.0 (#167), Nova 2 Lite: 27.5 (#118)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1364 |
| Kagi LLM Benchmark | 50% | — |
| CritPt | 0% | — |
| DTBench | 75.5% | — |
| LMCA | 26.1% | — |
| Surface Evolver Bench | 26.9% | — |
Math Nova 2 Lite leads
Mistral Medium: 28.1 (#245), Nova 2 Lite: 37.5 (#156)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Math | 1408 | 1359 |
| OTIS Mock AIME 2024-2025 | 32.2% | — |
| ProofBench | 9% | — |
| MATH Level 5 | 81.6% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge Nova 2 Lite leads
Mistral Medium: 25.0 (#265), Nova 2 Lite: 43.0 (#94)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| Vectara Hallucination Rate | 22.7% | 5.1% |
| LMArena Expert | 1408 | 1358 |
| GPQA Diamond | 59.5% | — |
| Humanity's Last Exam | 4.5% | — |
Multimodal Not comparable
Mistral Medium: 35.3 (#88), Nova 2 Lite: —
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Vision | 1172 | — |
Multilingual Mistral Medium leads
Mistral Medium: 52.1 (#91), Nova 2 Lite: 47.1 (#153)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Non-English | 1408 | 1337 |
| LMArena Chinese | 1447 | 1364 |
| LMArena French | 1459 | 1381 |
| LMArena German | 1432 | 1343 |
| LMArena Japanese | 1378 | 1271 |
| LMArena Korean | 1380 | 1284 |
| LMArena Russian | 1411 | 1343 |
| LMArena Spanish | 1433 | 1373 |
Instruction Following Mistral Medium leads
Mistral Medium: 73.7 (#116), Nova 2 Lite: 70.5 (#161)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Instruction Following | 1398 | 1335 |
Long Context Mistral Medium leads
Mistral Medium: 42.9 (#114), Nova 2 Lite: 40.6 (#150)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Longer Query | 1406 | 1335 |
Writing & Preference Mistral Medium leads
Mistral Medium: 60.0 (#103), Nova 2 Lite: 53.9 (#154)
| Benchmark | Mistral Medium | Nova 2 Lite |
|---|---|---|
| LMArena Text | 1424 | 1362 |
| LMArena Creative Writing | 1391 | 1291 |
| LMArena Multi-Turn | 1418 | 1338 |
| Short-Story Creative Writing | 77.3% | — |
Frequently asked questions
Is Mistral Medium better than Nova 2 Lite?
Nova 2 Lite is the stronger model overall, scoring 39.7 to 36.3 on the Noometry Index.
Which is cheaper, Mistral Medium or Nova 2 Lite?
Nova 2 Lite is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Mistral Medium lists at $1.50 and $7.50.
Is Mistral Medium or Nova 2 Lite better for coding?
Nova 2 Lite scores higher on coding benchmarks: 40.7 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
Nova 2 Lite does, with 1M tokens against 262K.
How many benchmarks do Mistral Medium and Nova 2 Lite share?
19 benchmarks have published results for both models. Mistral Medium has 36 scored results on Noometry and Nova 2 Lite has 19.