Model comparison
Mistral Small vs Qwen3.8 27B
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 33.4 on the Noometry Index. Mistral Small costs 4.2× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Mistral AI
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Mistral Small scores higher in 0 categories and Qwen3.8 27B in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 27B leads 41.0 to 19.8.
- The biggest single-benchmark swing is LMCA: 20.6% for Mistral Small and 41.4% for Qwen3.8 27B.
- Mistral Small is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.99 / $1.49 for Qwen3.8 27B.
Side by side
| Mistral Small | Qwen3.8 27B | |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 33.4 | 46.0 |
| Released | 2024-02-26 | 2026-08-14 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 256K | 33K |
| Input $ / M tokens | $0.15 | $0.99 |
| Output $ / M tokens | $0.60 | $1.49 |
| Results tracked | 39 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 27B leads
Mistral Small: 34.0 (#247), Qwen3.8 27B: 50.5 (#44)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| SciCode | 26.5% | 46.6% |
| LMArena Coding | 1362 | 1482 |
| LMArena WebDev | — | 1593 |
| BigCodeBench Instruct | 36.1% | — |
| LiveBench Coding | 36.2% | — |
| BigCodeBench Complete | 46.6% | — |
| ALE-Bench | 497.62 | — |
Agentic & Tool Use Qwen3.8 27B leads
Mistral Small: 28.1 (#93), Qwen3.8 27B: 32.9 (#57)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| APEX-Agents | — | 47.5% |
| Berkeley Function Calling Leaderboard | 37.1% | — |
Reasoning Qwen3.8 27B leads
Mistral Small: 19.8 (#250), Qwen3.8 27B: 41.0 (#54)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| CritPt | 0% | 5.4% |
| LMArena Hard Prompts | 1335 | 1460 |
| DTBench | 70.9% | 88% |
| LMCA | 20.6% | 41.4% |
| ARC-AGI-2 | — | 42.4% |
| Kagi LLM Benchmark | 37.8% | — |
| NYT Connections (extended) | — | 54.5% |
| ARC-AGI-1 | — | 87.5% |
| LiveBench Reasoning | 44.8% | — |
| LiveBench Data Analysis | 53.7% | — |
| Surface Evolver Bench | — | 45% |
| Epoch Capabilities Index | — | 149.38 |
| LiveBench | 44% | — |
Math Qwen3.8 27B leads
Mistral Small: 16.4 (#293), Qwen3.8 27B: 37.1 (#161)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Math | 1341 | 1456 |
| OTIS Mock AIME 2024-2025 | 5.8% | — |
| ProofBench | — | 16% |
| LiveBench Math | 39.9% | — |
| MATH Level 5 | 46.8% | — |
Knowledge Qwen3.8 27B leads
Mistral Small: 31.0 (#222), Qwen3.8 27B: 41.6 (#109)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Expert | 1291 | 1482 |
| GPQA Diamond | 47.5% | — |
| Vectara Hallucination Rate | 5.1% | — |
| MMLU | 68.7% | — |
Multimodal Qwen3.8 27B leads
Mistral Small: 33.5 (#96), Qwen3.8 27B: 41.3 (#37)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Vision | 1142 | 1271 |
Multilingual Qwen3.8 27B leads
Mistral Small: 45.5 (#169), Qwen3.8 27B: 53.7 (#60)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Non-English | 1315 | 1430 |
| LMArena Chinese | 1340 | 1504 |
| LMArena French | 1337 | 1465 |
| LMArena German | 1340 | 1438 |
| LMArena Japanese | 1275 | 1384 |
| LMArena Korean | 1259 | 1393 |
| LMArena Russian | 1324 | 1415 |
| LMArena Spanish | 1346 | 1448 |
Instruction Following Qwen3.8 27B leads
Mistral Small: 66.4 (#209), Qwen3.8 27B: 75.8 (#53)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Instruction Following | 1310 | 1439 |
| LiveBench Instruction Following | 63.7% | — |
Long Context Qwen3.8 27B leads
Mistral Small: 40.4 (#156), Qwen3.8 27B: 44.3 (#70)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Longer Query | 1327 | 1450 |
Writing & Preference Qwen3.8 27B leads
Mistral Small: 52.5 (#171), Qwen3.8 27B: 65.8 (#43)
| Benchmark | Mistral Small | Qwen3.8 27B |
|---|---|---|
| LMArena Text | 1338 | 1441 |
| LMArena Creative Writing | 1305 | 1384 |
| LMArena Multi-Turn | 1344 | 1441 |
| EQ-Bench Creative Writing | — | 1671 |
| LiveBench Language | 30.5% | — |
Frequently asked questions
Is Mistral Small better than Qwen3.8 27B?
Qwen3.8 27B is the stronger model overall, scoring 46.0 to 33.4 on the Noometry Index. Mistral Small costs 4.2× less per token, which makes it the better buy when Qwen3.8 27B's lead doesn't matter for your workload.
Which is cheaper, Mistral Small or Qwen3.8 27B?
Mistral Small is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Qwen3.8 27B lists at $0.99 and $1.49.
Is Mistral Small or Qwen3.8 27B better for coding?
Qwen3.8 27B scores higher on coding benchmarks: 50.5 versus 34.0 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Mistral Small and Qwen3.8 27B share?
22 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Qwen3.8 27B has 31.