Model comparison
DeepSeek V4.1 Flash vs Kimi K2 Thinking Turbo
DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 45.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
DeepSeekSummary
- They share 15 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 7 categories and Kimi K2 Thinking Turbo in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4.1 Flash leads 46.2 to 31.4.
Side by side
| DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 50.3 | 45.4 |
| Released | 2026-09-09 | 2025-11-06 |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 25 | 21 |
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: 52.8 (#29), Kimi K2 Thinking Turbo: 38.1 (#166)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena WebDev | 1620 | 1322 |
| LMArena Coding | 1511 | 1454 |
| SciCode | 51.9% | — |
| ALE-Bench | 1,092 | — |
Reasoning DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 46.2 (#39), Kimi K2 Thinking Turbo: 31.4 (#72)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Hard Prompts | 1485 | 1428 |
| NYT Connections (extended) | 89.6% | — |
| CritPt | 14.3% | — |
| Chess Puzzles | — | 20% |
| DTBench | 89.9% | — |
| LMCA | 47% | — |
| Surface Evolver Bench | 46.3% | — |
| Epoch Capabilities Index | 155.01 | — |
Math DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 48.4 (#52), Kimi K2 Thinking Turbo: 45.5 (#62)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Math | 1493 | 1429 |
| OTIS Mock AIME 2024-2025 | — | 83.1% |
| ProofBench | 54% | — |
Knowledge Kimi K2 Thinking Turbo leads
DeepSeek V4.1 Flash: 41.6 (#97), Kimi K2 Thinking Turbo: 50.9 (#64)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Expert | 1505 | 1439 |
| GPQA Diamond | — | 84.2% |
Multilingual DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 55.5 (#29), Kimi K2 Thinking Turbo: 51.6 (#104)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Non-English | 1452 | 1398 |
| LMArena Chinese | 1506 | 1457 |
| LMArena French | 1448 | 1425 |
| LMArena Russian | 1473 | 1390 |
| LMArena Spanish | 1499 | 1407 |
| LMArena German | — | 1390 |
| LMArena Japanese | — | 1356 |
| LMArena Korean | — | 1329 |
Instruction Following DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 77.6 (#24), Kimi K2 Thinking Turbo: 74.2 (#103)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Instruction Following | 1475 | 1403 |
Long Context DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 45.2 (#46), Kimi K2 Thinking Turbo: 43.2 (#99)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Longer Query | 1476 | 1415 |
Writing & Preference DeepSeek V4.1 Flash leads
DeepSeek V4.1 Flash: 65.8 (#43), Kimi K2 Thinking Turbo: 60.2 (#99)
| Benchmark | DeepSeek V4.1 Flash | Kimi K2 Thinking Turbo |
|---|---|---|
| LMArena Text | 1465 | 1415 |
| LMArena Creative Writing | 1436 | 1373 |
| LMArena Multi-Turn | 1460 | 1414 |
| EQ-Bench Creative Writing | 1540 | — |
Frequently asked questions
Is DeepSeek V4.1 Flash better than Kimi K2 Thinking Turbo?
DeepSeek V4.1 Flash is the stronger model overall, scoring 50.3 to 45.4 on the Noometry Index.
Is DeepSeek V4.1 Flash or Kimi K2 Thinking Turbo better for coding?
DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.8 versus 38.1 in the Noometry coding category.
How many benchmarks do DeepSeek V4.1 Flash and Kimi K2 Thinking Turbo share?
15 benchmarks have published results for both models. DeepSeek V4.1 Flash has 25 scored results on Noometry and Kimi K2 Thinking Turbo has 21.