Model comparison
DeepSeek-V3.1-Terminus vs Kimi K2 Thinking Turbo
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 43.1 on the Noometry Index.
Last verified . 10 shared benchmarks.
DeepSeekSummary
- They share 10 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 5 categories and Kimi K2 Thinking Turbo in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2 Thinking Turbo leads 47.4 to 38.5.
Side by side
| DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 43.1 | 45.8 |
| Released | 2025-09-22 | 2025-11-06 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 147K | — |
| Input $ / M tokens | $0.27 | — |
| Output $ / M tokens | $1 | — |
| Results tracked | 16 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Kimi K2 Thinking Turbo: 38.4 (#178)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Coding | 1426 | 1454 |
| LMArena WebDev | — | 1322 |
| SciCode | 40.6% | — |
| ALE-Bench | 745.17 | — |
Reasoning Kimi K2 Thinking Turbo leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Kimi K2 Thinking Turbo: 32.5 (#75)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Hard Prompts | 1426 | 1428 |
| Kagi LLM Benchmark | 57.4% | — |
| CritPt | 1.7% | — |
| Chess Puzzles | — | 20% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
Math Kimi K2 Thinking Turbo leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Kimi K2 Thinking Turbo: 47.4 (#68)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Math | 1402 | 1429 |
| OTIS Mock AIME 2024-2025 | — | 83.1% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2 Thinking Turbo: 50.9 (#69)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| GPQA Diamond | — | 84.2% |
| LMArena Expert | — | 1439 |
Multilingual Too close to call
DeepSeek-V3.1-Terminus: 52.1 (#92), Kimi K2 Thinking Turbo: 51.4 (#109)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Non-English | 1407 | 1398 |
| LMArena Russian | 1436 | 1391 |
| LMArena Chinese | — | 1456 |
| LMArena French | — | 1425 |
| LMArena German | — | 1390 |
| LMArena Japanese | — | 1357 |
| LMArena Korean | — | 1330 |
| LMArena Spanish | — | 1406 |
Instruction Following Too close to call
DeepSeek-V3.1-Terminus: 74.0 (#106), Kimi K2 Thinking Turbo: 74.0 (#109)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Instruction Following | 1404 | 1403 |
Long Context Too close to call
DeepSeek-V3.1-Terminus: 43.4 (#97), Kimi K2 Thinking Turbo: 43.2 (#102)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Longer Query | 1421 | 1415 |
Writing & Preference DeepSeek-V3.1-Terminus leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Kimi K2 Thinking Turbo: 60.0 (#104)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Text | 1419 | 1415 |
| LMArena Creative Writing | 1403 | 1374 |
| LMArena Multi-Turn | 1411 | 1414 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Kimi K2 Thinking Turbo?
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 43.1 on the Noometry Index.
Is DeepSeek-V3.1-Terminus or Kimi K2 Thinking Turbo better for coding?
DeepSeek-V3.1-Terminus scores higher on coding benchmarks: 42.0 versus 38.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1-Terminus and Kimi K2 Thinking Turbo share?
10 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Kimi K2 Thinking Turbo has 21.