Model comparison
DeepSeek V4.1 Flash vs Kimi K2.5
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 48.1 on the Noometry Index.
Last verified . 27 shared benchmarks.
DeepSeekSummary
- They share 27 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 7 categories and Kimi K2.5 in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 50.2 to 31.2.
- The biggest single-benchmark swing is NYT Connections (extended): 89.6% for DeepSeek V4.1 Flash and 69.9% for Kimi K2.5.
- DeepSeek V4.1 Flash is cheaper at $0.15 / $0.60 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- DeepSeek V4.1 Flash accepts more context: 1M tokens versus 262K.
Side by side
| DeepSeek V4.1 Flash | Kimi K2.5 | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 52.8 | 48.1 |
| Released | 2026-09-09 | 2026-01-27 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 393K | 262K |
| Input $ / M tokens | $0.15 | $0.45 |
| Output $ / M tokens | $0.60 | $2.25 |
| Results tracked | 37 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 52.9 (#32), Kimi K2.5: 48.8 (#53)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| LMArena WebDev | 1619 | 1437 |
| SciCode | 51.9% | 49% |
| LMArena Coding | 1506 | 1474 |
| ALE-Bench | 1,092 | 821.65 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 70.8% |
| SWE-bench Multilingual | — | 67.3% |
| WeirdML | — | 45.6% |
Agentic & Tool Use Kimi K2.5 leads
DeepSeek V4.1 Flash: 31.2 (#69), Kimi K2.5: 34.2 (#48)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | — | 43.2% |
| APEX-Agents | 39.5% | — |
| OSWorld | — | 63.3% |
| GDP.pdf | 19.8% | — |
| Vending-Bench 2 | — | 1,198 |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 50.2 (#36), Kimi K2.5: 31.2 (#80)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| NYT Connections (extended) | 89.6% | 69.9% |
| CritPt | 14.3% | 3.1% |
| LMArena Hard Prompts | 1483 | 1453 |
| Epoch Capabilities Index | 154.9 | 148.03 |
| ARC-AGI-2 | — | 11.8% |
| SimpleBench | — | 46.8% |
| Kagi LLM Benchmark | — | 78.5% |
| ARC-AGI-1 | — | 65.3% |
| Chess Puzzles | — | 12% |
| EnigmaEval | — | 3.4% |
| Thematic Generalization | — | 69.4% |
| Mystery Game Puzzles | 43% | — |
| DTBench | 89.9% | — |
| LMCA | 47% | — |
| Surface Evolver Bench | 46.3% | — |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 66.7 (#25), Kimi K2.5: 51.8 (#53)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3% | 92.2% |
| LMArena Math | 1477 | 1470 |
| FrontierMath (Tiers 1-3) | 67.4% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | — | 62.3% |
| ProofBench | 54% | — |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 57.9 (#38), Kimi K2.5: 53.6 (#56)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | 89.8% | 87.6% |
| LMArena Expert | 1506 | 1466 |
| Humanity's Last Exam | — | 24.4% |
| SimpleQA Verified | — | 34.3% |
| Vectara Hallucination Rate | — | 14.2% |
Multimodal Kimi K2.5 leads
DeepSeek V4.1 Flash: 39.1 (#61), Kimi K2.5: 41.1 (#39)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| LMArena Vision | 1277 | 1269 |
| Furniture Assembly | 34.2% | — |
| LMArena Document | — | 1430 |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.0 (#35), Kimi K2.5: 53.9 (#53)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | 1448 | 1433 |
| LMArena Chinese | 1497 | 1495 |
| LMArena French | 1452 | 1454 |
| LMArena German | 1484 | 1441 |
| LMArena Japanese | 1412 | 1421 |
| LMArena Korean | 1452 | 1410 |
| LMArena Russian | 1471 | 1435 |
| LMArena Spanish | 1459 | 1450 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.3 (#26), Kimi K2.5: 75.3 (#64)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | 1474 | 1431 |
Long Context Kimi K2.5 leads
DeepSeek V4.1 Flash: 45.2 (#47), Kimi K2.5: 52.1 (#7)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| LMArena Longer Query | 1475 | 1445 |
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
Writing & Preference Too close to call
DeepSeek V4.1 Flash: 65.4 (#48), Kimi K2.5: 65.1 (#53)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2.5 |
|---|---|---|
| LMArena Text | 1462 | 1445 |
| LMArena Creative Writing | 1435 | 1423 |
| EQ-Bench Creative Writing | 1540 | 1579 |
| LMArena Multi-Turn | 1457 | 1444 |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Kimi K2.5?
DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 48.1 on the Noometry Index.
Which is cheaper, DeepSeek V4.1 Flash or Kimi K2.5?
DeepSeek V4.1 Flash is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is DeepSeek V4.1 Flash or Kimi K2.5 better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 48.8 in the Noometry coding category.
Which has the bigger context window?
DeepSeek V4.1 Flash does, with 1M tokens against 262K.
How many benchmarks do DeepSeek V4.1 Flash and Kimi K2.5 share?
27 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Kimi K2.5 has 51.