Model comparison
DeepSeek-V3.1-Terminus vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.7 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 3.8× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
DeepSeekSummary
- They share 15 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and Kimi K2.6 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 57.0 to 38.5.
- The biggest single-benchmark swing is SciCode: 40.6% for DeepSeek-V3.1-Terminus and 53.5% for Kimi K2.6.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Kimi K2.6 accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.1-Terminus | Kimi K2.6 | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 43.1 | 47.7 |
| Released | 2025-09-22 | 2026-04-20 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 147K | 262K |
| Input $ / M tokens | $0.27 | $0.95 |
| Output $ / M tokens | $1 | $4 |
| Results tracked | 16 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Kimi K2.6: 50.7 (#43)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| SciCode | 40.6% | 53.5% |
| LMArena Coding | 1426 | 1488 |
| ALE-Bench | 745.17 | 1,093 |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1509 |
| WeirdML | — | 55.9% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2.6: 21.9 (#137)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| OSWorld 2.0 | — | 4.6% |
| ExploitBench | — | 18.4% |
| GBAEval | — | 0.9% |
| GDP.pdf | — | 12% |
| Vending-Bench 2 | — | 6,205 |
Reasoning Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Kimi K2.6: 40.5 (#55)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| CritPt | 1.7% | 8% |
| LMArena Hard Prompts | 1426 | 1470 |
| DTBench | 81.3% | 90.9% |
| LMCA | 28.6% | 37.3% |
| Kagi LLM Benchmark | 57.4% | — |
| NYT Connections (extended) | — | 87.2% |
| Chess Puzzles | — | 26% |
| EBR-Bench | — | 2.4% |
| Mystery Game Puzzles | — | 18% |
| Epoch Capabilities Index | — | 151.05 |
Math Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Kimi K2.6: 57.0 (#41)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| LMArena Math | 1402 | 1475 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| OTIS Mock AIME 2024-2025 | — | 96.1% |
| ProofBench | — | 16% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2.6: 54.0 (#54)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | — | 90.8% |
| SimpleQA Verified | — | 34.9% |
| Vectara Hallucination Rate | — | 10.8% |
| LMArena Expert | — | 1491 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2.6: 31.6 (#103)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| LMArena Vision | — | 1283 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Kimi K2.6: 54.9 (#37)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1407 | 1446 |
| LMArena Russian | 1436 | 1446 |
| LMArena Chinese | — | 1521 |
| LMArena French | — | 1471 |
| LMArena German | — | 1450 |
| LMArena Japanese | — | 1443 |
| LMArena Korean | — | 1427 |
| LMArena Spanish | — | 1464 |
Instruction Following Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Kimi K2.6: 76.3 (#43)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1451 |
Long Context Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Kimi K2.6: 44.9 (#52)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1421 | 1468 |
Writing & Preference Kimi K2.6 leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Kimi K2.6: 68.5 (#26)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1419 | 1455 |
| LMArena Creative Writing | 1403 | 1434 |
| LMArena Multi-Turn | 1411 | 1453 |
| EQ-Bench Creative Writing | — | 1725 |
| EQ-Bench 4 | — | 1202 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.7 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 3.8× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1-Terminus or Kimi K2.6?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is DeepSeek-V3.1-Terminus or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 42.0 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and Kimi K2.6 share?
15 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Kimi K2.6 has 51.