Model comparison
Mistral Large 3 vs Qwen3.5 397B-A17B
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 38.6 on the Noometry Index. Mistral Large 3 costs 3.6× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Last verified . 23 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Mistral Large 3 scores higher in 0 categories and Qwen3.5 397B-A17B in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.5 397B-A17B leads 33.6 to 14.6.
- The biggest single-benchmark swing is NYT Connections (extended): 7.5% for Mistral Large 3 and 58.9% for Qwen3.5 397B-A17B.
- Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $0.60 / $3.60 for Qwen3.5 397B-A17B.
Side by side
| Mistral Large 3 | Qwen3.5 397B-A17B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 38.6 | 45.7 |
| Released | 2025-12-02 | 2026-02-01 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 8K | 66K |
| Input $ / M tokens | $0.25 | $0.60 |
| Output $ / M tokens | $0.75 | $3.60 |
| Results tracked | 24 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.5 397B-A17B leads
Mistral Large 3: 34.1 (#230), Qwen3.5 397B-A17B: 41.7 (#108)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena WebDev | 1230 | 1400 |
| LMArena Coding | 1447 | 1464 |
Agentic & Tool Use Not comparable
Mistral Large 3: —, Qwen3.5 397B-A17B: 33.2 (#51)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| APEX-Agents | — | 24.9% |
| τ²-bench Airline | — | 81.5% |
| τ²-bench Banking | — | 9.8% |
| τ²-bench Retail | — | 84.4% |
| τ²-bench Telecom | — | 97.8% |
Reasoning Qwen3.5 397B-A17B leads
Mistral Large 3: 14.6 (#314), Qwen3.5 397B-A17B: 33.6 (#62)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| Kagi LLM Benchmark | 50.9% | 73.7% |
| NYT Connections (extended) | 7.5% | 58.9% |
| Thematic Generalization | 23% | 65.1% |
| LMArena Hard Prompts | 1429 | 1448 |
| Chess Puzzles | — | 13% |
| Mystery Game Puzzles | — | 18% |
| DTBench | — | 87.5% |
| LMCA | — | 37.9% |
| Epoch Capabilities Index | — | 146.64 |
Math Qwen3.5 397B-A17B leads
Mistral Large 3: 36.0 (#131), Qwen3.5 397B-A17B: 44.9 (#64)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Math | 1415 | 1455 |
| FrontierMath (Tiers 1-3) | — | 31.2% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
Knowledge Qwen3.5 397B-A17B leads
Mistral Large 3: 35.5 (#173), Qwen3.5 397B-A17B: 53.3 (#53)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Expert | 1421 | 1461 |
| GPQA Diamond | — | 86.4% |
| Vectara Hallucination Rate | 14.5% | — |
Multimodal Qwen3.5 397B-A17B leads
Mistral Large 3: 37.6 (#65), Qwen3.5 397B-A17B: 40.0 (#42)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Vision | 1222 | 1262 |
Multilingual Qwen3.5 397B-A17B leads
Mistral Large 3: 52.6 (#79), Qwen3.5 397B-A17B: 54.0 (#56)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Non-English | 1412 | 1431 |
| LMArena Chinese | 1448 | 1501 |
| LMArena French | 1456 | 1461 |
| LMArena German | 1438 | 1447 |
| LMArena Japanese | 1399 | 1427 |
| LMArena Korean | 1384 | 1382 |
| LMArena Russian | 1412 | 1429 |
| LMArena Spanish | 1437 | 1443 |
Instruction Following Qwen3.5 397B-A17B leads
Mistral Large 3: 74.1 (#104), Qwen3.5 397B-A17B: 75.2 (#73)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Instruction Following | 1402 | 1424 |
Long Context Too close to call
Mistral Large 3: 43.1 (#102), Qwen3.5 397B-A17B: 44.1 (#71)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Longer Query | 1413 | 1442 |
Writing & Preference Qwen3.5 397B-A17B leads
Mistral Large 3: 60.2 (#100), Qwen3.5 397B-A17B: 62.5 (#74)
| Benchmark | Mistral Large 3 | Qwen3.5 397B-A17B |
|---|---|---|
| LMArena Text | 1427 | 1438 |
| LMArena Creative Writing | 1385 | 1401 |
| EQ-Bench Creative Writing | 1412 | 1478 |
| LMArena Multi-Turn | 1429 | 1447 |
Frequently asked questions
Is Mistral Large 3 better than Qwen3.5 397B-A17B?
Qwen3.5 397B-A17B is the stronger model overall, scoring 45.7 to 38.6 on the Noometry Index. Mistral Large 3 costs 3.6× less per token, which makes it the better buy when Qwen3.5 397B-A17B's lead doesn't matter for your workload.
Which is cheaper, Mistral Large 3 or Qwen3.5 397B-A17B?
Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Qwen3.5 397B-A17B lists at $0.60 and $3.60.
Is Mistral Large 3 or Qwen3.5 397B-A17B better for coding?
Qwen3.5 397B-A17B scores higher on coding benchmarks: 41.7 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Mistral Large 3 and Qwen3.5 397B-A17B share?
23 benchmarks have published results for both models. Mistral Large 3 has 24 scored results on Noometry and Qwen3.5 397B-A17B has 36.