Model comparison
DeepSeek-V3.1 vs GPT-5 Mini
DeepSeek-V3.1 and GPT-5 Mini score almost the same on the Noometry Index (42.2 vs 41.7), so choose on price, context window or the category you care about most.
Last verified . 26 shared benchmarks.
DeepSeekSummary
- They share 26 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 3 categories and GPT-5 Mini in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 36.1.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 53.2% for DeepSeek-V3.1 and 70.3% for GPT-5 Mini.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
- GPT-5 Mini accepts more context: 400K tokens versus 164K.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | GPT-5 Mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.2 | 41.7 |
| Released | 2025-08-21 | 2025-08-07 |
| Weights | Open | Proprietary |
| Context window | 164K | 400K |
| Max output | 8K | 128K |
| Input $ / M tokens | $0.25 | $0.25 |
| Output $ / M tokens | $0.95 | $2 |
| Results tracked | 27 | 60 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.1: 39.8 (#140), GPT-5 Mini: 40.0 (#134)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| WeirdML | 38.4% | 52.7% |
| LMArena Coding | 1417 | 1406 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| SWE-bench Multilingual | — | 39.7% |
| SciCode | — | 39.2% |
| ALE-Bench | — | 799.77 |
| AlgoTune | — | 1.38 |
Agentic & Tool Use Not comparable
DeepSeek-V3.1: —, GPT-5 Mini: 31.0 (#64)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| Vending-Bench 2 | — | -31.18 |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.3 (#109), GPT-5 Mini: 24.1 (#153)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| Kagi LLM Benchmark | 53.2% | 70.3% |
| LMArena Hard Prompts | 1417 | 1380 |
| DTBench | 82.7% | 80.5% |
| LMCA | 24.3% | 34.2% |
| Epoch Capabilities Index | 139.92 | 145.51 |
| ForecastBench | 58 | 61 |
| ARC-AGI-2 | — | 4.4% |
| SimpleBench | 40% | — |
| ARC-AGI-1 | — | 54.3% |
| CritPt | — | 0% |
| Chess Puzzles | — | 30% |
| EnigmaEval | — | 8.2% |
| Mystery Game Puzzles | — | 10% |
Math GPT-5 Mini leads
DeepSeek-V3.1: 36.1 (#125), GPT-5 Mini: 46.3 (#59)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| LMArena Math | 1420 | 1377 |
| FrontierMath (Tiers 1-3) | — | 46.7% |
| FrontierMath Tier 4 | — | 12.2% |
| OTIS Mock AIME 2024-2025 | — | 86.7% |
| ProofBench | — | 9% |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |
Knowledge GPT-5 Mini leads
DeepSeek-V3.1: 43.0 (#86), GPT-5 Mini: 45.5 (#80)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | 12.9% |
| LMArena Expert | 1403 | 1377 |
| GPQA Diamond | — | 75% |
| Humanity's Last Exam | — | 19.4% |
| SimpleQA Verified | — | 21.6% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| GPQA (HELM) | — | 75.6% |
Multimodal Not comparable
DeepSeek-V3.1: —, GPT-5 Mini: 35.0 (#80)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| LMArena Vision | — | 1202 |
| VPCT | — | 40.2% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.8 (#100), GPT-5 Mini: 49.1 (#131)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1400 | 1363 |
| LMArena Chinese | 1470 | 1387 |
| LMArena French | 1447 | 1385 |
| LMArena German | 1412 | 1366 |
| LMArena Japanese | 1374 | 1342 |
| LMArena Korean | 1337 | 1306 |
| LMArena Russian | 1405 | 1362 |
| LMArena Spanish | 1429 | 1355 |
Instruction Following GPT-5 Mini leads
DeepSeek-V3.1: 74.1 (#105), GPT-5 Mini: 76.4 (#43)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1400 | 1357 |
| IFEval | — | 92.7% |
Long Context GPT-5 Mini leads
DeepSeek-V3.1: 36.3 (#227), GPT-5 Mini: 41.9 (#127)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| Fiction.LiveBench | 52.8% | 69.4% |
| LMArena Longer Query | 1422 | 1355 |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.5 (#93), GPT-5 Mini: 55.4 (#143)
| Benchmark | DeepSeek-V3.1 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1420 | 1373 |
| LMArena Creative Writing | 1401 | 1325 |
| EQ-Bench Creative Writing | 1436 | 1313 |
| LMArena Multi-Turn | 1408 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |
Frequently asked questions
Is DeepSeek-V3.1 better than GPT-5 Mini?
DeepSeek-V3.1 and GPT-5 Mini score almost the same on the Noometry Index (42.2 vs 41.7), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.1 or GPT-5 Mini?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; GPT-5 Mini lists at $0.25 and $2.
Is DeepSeek-V3.1 or GPT-5 Mini better for coding?
They score almost the same on coding (39.8 vs 40.0); test both on your own repository before choosing.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 164K.
How many benchmarks do DeepSeek-V3.1 and GPT-5 Mini share?
26 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and GPT-5 Mini has 60.