Model comparison
DeepSeek-V3.1-Terminus vs o3-mini
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 36.7 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeekSummary
- They share 14 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 6 categories and o3-mini in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1-Terminus leads 61.3 to 50.4.
- The biggest single-benchmark swing is DTBench: 81.3% for DeepSeek-V3.1-Terminus and 68.8% for o3-mini.
- DeepSeek-V3.1-Terminus is cheaper at $0.30 / $1 per million input/output tokens, against $1.10 / $4.40 for o3-mini.
- o3-mini accepts more context: 200K tokens versus 164K.
- DeepSeek-V3.1-Terminus has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1-Terminus | o3-mini | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 42.6 | 36.7 |
| Released | 2025-09-22 | 2024-12-20 |
| Weights | Open | Proprietary |
| Context window | 164K | 200K |
| Max output | 66K | 100K |
| Input $ / M tokens | $0.30 | $1.10 |
| Output $ / M tokens | $1 | $4.40 |
| Results tracked | 16 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek-V3.1-Terminus: 41.5 (#114), o3-mini: 41.4 (#116)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| SciCode | 40.6% | 39.8% |
| LMArena Coding | 1426 | 1378 |
| Aider Polyglot | — | 60.4% |
| GSO | — | 1.3% |
| WeirdML | — | 43.7% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
| ALE-Bench | 745.17 | — |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, o3-mini: 29.3 (#79)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| Cybench | — | 22.5% |
Reasoning DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 25.7 (#126), o3-mini: 16.3 (#295)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| CritPt | 1.7% | 0.3% |
| LMArena Hard Prompts | 1427 | 1366 |
| DTBench | 81.3% | 68.8% |
| LMCA | 28.6% | 19% |
| ARC-AGI-2 | — | 3% |
| SimpleBench | — | 22.8% |
| Kagi LLM Benchmark | 57.4% | — |
| ARC-AGI-1 | — | 34.5% |
| Chess Puzzles | — | 17% |
| LiveBench Reasoning | — | 89.6% |
| Mystery Game Puzzles | — | 7% |
| LiveBench Data Analysis | — | 70.6% |
| Epoch Capabilities Index | — | 140.35 |
| ForecastBench | — | 59.6 |
| LiveBench | — | 75.9% |
Math DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 35.7 (#138), o3-mini: 27.6 (#236)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| LMArena Math | 1402 | 1396 |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| OTIS Mock AIME 2024-2025 | — | 76.9% |
| LiveBench Math | — | 77.3% |
| MATH Level 5 | — | 96.5% |
| FrontierMath (Feb 2025 set) | — | 12.4% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, o3-mini: 38.3 (#135)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| GPQA Diamond | — | 77% |
| SimpleQA Verified | — | 15.3% |
| Confabulations | — | 17.9% |
| LMArena Expert | — | 1364 |
Multilingual DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 52.3 (#87), o3-mini: 45.9 (#159)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| LMArena Non-English | 1407 | 1319 |
| LMArena Russian | 1434 | 1304 |
| LMArena Chinese | — | 1379 |
| LMArena French | — | 1333 |
| LMArena German | — | 1303 |
| LMArena Japanese | — | 1286 |
| LMArena Korean | — | 1314 |
| LMArena Spanish | — | 1321 |
Instruction Following Too close to call
DeepSeek-V3.1-Terminus: 74.4 (#96), o3-mini: 75.3 (#67)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1406 | 1338 |
| LiveBench Instruction Following | — | 84.4% |
Long Context DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 43.5 (#88), o3-mini: 33.8 (#251)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| LMArena Longer Query | 1426 | 1343 |
| Fiction.LiveBench | — | 50% |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.3 (#87), o3-mini: 50.4 (#177)
| Benchmark | DeepSeek-V3.1-Terminus | o3-mini |
|---|---|---|
| LMArena Text | 1420 | 1337 |
| LMArena Creative Writing | 1404 | 1286 |
| LMArena Multi-Turn | 1411 | 1320 |
| Short-Story Creative Writing | — | 61.7% |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than o3-mini?
DeepSeek-V3.1-Terminus is the stronger model overall, scoring 42.6 to 36.7 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1-Terminus or o3-mini?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.30 per million input tokens and $1 per million output tokens; o3-mini lists at $1.10 and $4.40.
Is DeepSeek-V3.1-Terminus or o3-mini better for coding?
They score almost the same on coding (41.5 vs 41.4); test both on your own repository before choosing.
Which has the bigger context window?
o3-mini does, with 200K tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and o3-mini share?
14 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and o3-mini has 51.