Model comparison
DeepSeek-V3.1-Terminus vs Kimi K2.5
Kimi K2.5 is the stronger model overall, scoring 48.1 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 2.0× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Last verified . 14 shared benchmarks.
DeepSeekSummary
- They share 14 benchmarks with published results for both. DeepSeek-V3.1-Terminus scores higher in 0 categories and Kimi K2.5 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 leads 51.8 to 38.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.4% for DeepSeek-V3.1-Terminus and 78.5% for Kimi K2.5.
- DeepSeek-V3.1-Terminus is cheaper at $0.27 / $1 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- Kimi K2.5 accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.1-Terminus | Kimi K2.5 | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 43.1 | 48.1 |
| Released | 2025-09-22 | 2026-01-27 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 147K | 262K |
| Input $ / M tokens | $0.27 | $0.45 |
| Output $ / M tokens | $1 | $2.25 |
| Results tracked | 16 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 42.0 (#113), Kimi K2.5: 48.8 (#53)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| SciCode | 40.6% | 49% |
| LMArena Coding | 1426 | 1474 |
| ALE-Bench | 745.17 | 821.65 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 70.8% |
| LMArena WebDev | — | 1437 |
| SWE-bench Multilingual | — | 67.3% |
| WeirdML | — | 45.6% |
Agentic & Tool Use Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2.5: 34.2 (#48)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | — | 43.2% |
| OSWorld | — | 63.3% |
| Vending-Bench 2 | — | 1,198 |
Reasoning Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 26.4 (#133), Kimi K2.5: 31.2 (#80)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| Kagi LLM Benchmark | 57.4% | 78.5% |
| CritPt | 1.7% | 3.1% |
| LMArena Hard Prompts | 1426 | 1453 |
| ARC-AGI-2 | — | 11.8% |
| SimpleBench | — | 46.8% |
| NYT Connections (extended) | — | 69.9% |
| ARC-AGI-1 | — | 65.3% |
| Chess Puzzles | — | 12% |
| EnigmaEval | — | 3.4% |
| Thematic Generalization | — | 69.4% |
| DTBench | 81.3% | — |
| LMCA | 28.6% | — |
| Epoch Capabilities Index | — | 148.03 |
Math Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 38.5 (#137), Kimi K2.5: 51.8 (#53)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| LMArena Math | 1402 | 1470 |
| MathArena Final-Answer Competitions | — | 62.3% |
| OTIS Mock AIME 2024-2025 | — | 92.2% |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2.5: 53.6 (#56)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | — | 87.6% |
| Humanity's Last Exam | — | 24.4% |
| SimpleQA Verified | — | 34.3% |
| Vectara Hallucination Rate | — | 14.2% |
| LMArena Expert | — | 1466 |
Multimodal Not comparable
DeepSeek-V3.1-Terminus: —, Kimi K2.5: 41.1 (#39)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| LMArena Vision | — | 1269 |
| LMArena Document | — | 1430 |
Multilingual Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 52.1 (#92), Kimi K2.5: 53.9 (#53)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1407 | 1433 |
| LMArena Russian | 1436 | 1435 |
| LMArena Chinese | — | 1495 |
| LMArena French | — | 1454 |
| LMArena German | — | 1441 |
| LMArena Japanese | — | 1421 |
| LMArena Korean | — | 1410 |
| LMArena Spanish | — | 1450 |
Instruction Following Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 74.0 (#106), Kimi K2.5: 75.3 (#64)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1431 |
Long Context Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 43.4 (#97), Kimi K2.5: 52.1 (#7)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1421 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Kimi K2.5 leads
DeepSeek-V3.1-Terminus: 61.0 (#92), Kimi K2.5: 65.1 (#53)
| Benchmark | DeepSeek-V3.1-Terminus | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1419 | 1445 |
| LMArena Creative Writing | 1403 | 1423 |
| LMArena Multi-Turn | 1411 | 1444 |
| EQ-Bench Creative Writing | — | 1579 |
Frequently asked questions
Is DeepSeek-V3.1-Terminus better than Kimi K2.5?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 43.1 on the Noometry Index. DeepSeek-V3.1-Terminus costs 2.0× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Which is cheaper, DeepSeek-V3.1-Terminus or Kimi K2.5?
DeepSeek-V3.1-Terminus is cheaper. It lists at $0.27 per million input tokens and $1 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is DeepSeek-V3.1-Terminus or Kimi K2.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 42.0 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.1-Terminus and Kimi K2.5 share?
14 benchmarks have published results for both models. DeepSeek-V3.1-Terminus has 16 scored results on Noometry and Kimi K2.5 has 51.