Model comparison
gpt-oss-20b vs Mistral Medium 3.5
Mistral Medium 3.5 is the stronger model overall, scoring 40.2 to 32.5 on the Noometry Index. gpt-oss-20b costs 83× less per token, which makes it the better buy when Mistral Medium 3.5's lead doesn't matter for your workload.
Last verified . 17 shared benchmarks.
Mistral AISummary
- They share 17 benchmarks with published results for both. gpt-oss-20b scores higher in 3 categories and Mistral Medium 3.5 in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Mistral Medium 3.5 leads 58.5 to 35.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 53.2% for gpt-oss-20b and 41.4% for Mistral Medium 3.5.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $1.50 / $7.50 for Mistral Medium 3.5.
- Mistral Medium 3.5 accepts more context: 262K tokens versus 131K.
Side by side
| gpt-oss-20b | Mistral Medium 3.5 | |
|---|---|---|
| Provider | OpenAI | Mistral AI |
| Noometry Index | 32.5 | 40.2 |
| Released | 2025-08-05 | — |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 16K | 210K |
| Input $ / M tokens | $0.018 | $1.50 |
| Output $ / M tokens | $0.09 | $7.50 |
| Results tracked | 34 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
gpt-oss-20b: 37.6 (#192), Mistral Medium 3.5: 36.0 (#213)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Coding | 1306 | 1461 |
| LMArena WebDev | — | 1264 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Not comparable
gpt-oss-20b: 9.3 (#154), Mistral Medium 3.5: —
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| Terminal-Bench | 3.4% | — |
Reasoning gpt-oss-20b leads
gpt-oss-20b: 19.3 (#261), Mistral Medium 3.5: 17.3 (#295)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 41.4% |
| LMArena Hard Prompts | 1274 | 1436 |
| Epoch Capabilities Index | 137.82 | 141.35 |
| NYT Connections (extended) | — | 12.9% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| DTBench | 68% | — |
| LMCA | 14.5% | — |
Math Too close to call
gpt-oss-20b: 39.4 (#103), Mistral Medium 3.5: 39.1 (#113)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Math | 1317 | 1431 |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| Omni-MATH | 56.5% | — |
Knowledge Mistral Medium 3.5 leads
gpt-oss-20b: 34.6 (#195), Mistral Medium 3.5: 40.0 (#126)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Expert | 1258 | 1432 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
Multimodal Not comparable
gpt-oss-20b: —, Mistral Medium 3.5: 38.3 (#65)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Vision | — | 1223 |
Multilingual Mistral Medium 3.5 leads
gpt-oss-20b: 42.2 (#197), Mistral Medium 3.5: 51.9 (#100)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Non-English | 1268 | 1404 |
| LMArena Chinese | 1314 | 1442 |
| LMArena German | 1255 | 1451 |
| LMArena Korean | 1236 | 1385 |
| LMArena Russian | 1278 | 1395 |
| LMArena Spanish | 1267 | 1409 |
| LMArena French | — | 1448 |
| LMArena Japanese | 1244 | — |
Instruction Following Mistral Medium 3.5 leads
gpt-oss-20b: 61.8 (#240), Mistral Medium 3.5: 74.6 (#90)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Instruction Following | 1236 | 1415 |
| IFEval | 73.2% | — |
Long Context Mistral Medium 3.5 leads
gpt-oss-20b: 37.9 (#209), Mistral Medium 3.5: 43.2 (#103)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Longer Query | 1250 | 1415 |
Writing & Preference Mistral Medium 3.5 leads
gpt-oss-20b: 35.5 (#265), Mistral Medium 3.5: 58.5 (#117)
| Benchmark | gpt-oss-20b | Mistral Medium 3.5 |
|---|---|---|
| LMArena Text | 1287 | 1421 |
| LMArena Creative Writing | 1201 | 1374 |
| LMArena Multi-Turn | 1268 | 1423 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
| EQ-Bench 4 | — | 993 |
Frequently asked questions
Is gpt-oss-20b better than Mistral Medium 3.5?
Mistral Medium 3.5 is the stronger model overall, scoring 40.2 to 32.5 on the Noometry Index. gpt-oss-20b costs 83× less per token, which makes it the better buy when Mistral Medium 3.5's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-20b or Mistral Medium 3.5?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Mistral Medium 3.5 lists at $1.50 and $7.50.
Is gpt-oss-20b or Mistral Medium 3.5 better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 36.0 in the Noometry coding category.
Which has the bigger context window?
Mistral Medium 3.5 does, with 262K tokens against 131K.
How many benchmarks do gpt-oss-20b and Mistral Medium 3.5 share?
17 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Mistral Medium 3.5 has 22.