Model comparison
DeepSeek V4 Pro vs Kimi K2.5 Instant
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 43.6 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeekSummary
- They share 17 benchmarks with published results for both. DeepSeek V4 Pro scores higher in 8 categories and Kimi K2.5 Instant in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek V4 Pro leads 56.5 to 29.7.
Side by side
| DeepSeek V4 Pro | Kimi K2.5 Instant | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 54.3 | 43.6 |
| Released | 2026-04-24 | — |
| Weights | Open | Open |
| Context window | 1M | — |
| Max output | 393K | — |
| Input $ / M tokens | $0.66 | — |
| Output $ / M tokens | $1.98 | — |
| Results tracked | 48 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek V4 Pro leads
DeepSeek V4 Pro: 52.4 (#34), Kimi K2.5 Instant: 42.6 (#97)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena WebDev | 1582 | 1404 |
| LMArena Coding | 1470 | 1484 |
| SWE-bench Verified | 77.6% | — |
| FrontierCode | 28.6% | — |
| SciCode | 51% | — |
| WeirdML | 66.2% | — |
| ALE-Bench | 1,403 | — |
Agentic & Tool Use Not comparable
DeepSeek V4 Pro: 32.8 (#58), Kimi K2.5 Instant: —
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| APEX-Agents | 47.3% | — |
| Vending-Bench 2 | 3,285 | — |
Reasoning DeepSeek V4 Pro leads
DeepSeek V4 Pro: 56.5 (#24), Kimi K2.5 Instant: 29.7 (#90)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1443 |
| ARC-AGI-2 | 61.3% | — |
| Kagi LLM Benchmark | 53.5% | — |
| NYT Connections (extended) | 91.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | 18% | — |
| Chess Puzzles | 47% | — |
| Mystery Game Puzzles | 43% | — |
| DTBench | 93.9% | — |
| LMCA | 45.5% | — |
| Surface Evolver Bench | 40% | — |
| Epoch Capabilities Index | 155.31 | — |
| ForecastBench | 56.1 | — |
Math DeepSeek V4 Pro leads
DeepSeek V4 Pro: 64.8 (#30), Kimi K2.5 Instant: 39.4 (#105)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Math | 1455 | 1442 |
| FrontierMath (Tiers 1-3) | 64.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 76.6% | — |
| OTIS Mock AIME 2024-2025 | 98.6% | — |
| ProofBench | 50% | — |
Knowledge DeepSeek V4 Pro leads
DeepSeek V4 Pro: 59.5 (#31), Kimi K2.5 Instant: 40.2 (#123)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Expert | 1464 | 1440 |
| GPQA Diamond | 91.7% | — |
| SimpleQA Verified | 52.9% | — |
| Vectara Hallucination Rate | 8.6% | — |
Multimodal Not comparable
DeepSeek V4 Pro: —, Kimi K2.5 Instant: 40.2 (#50)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Vision | — | 1254 |
Multilingual DeepSeek V4 Pro leads
DeepSeek V4 Pro: 54.4 (#45), Kimi K2.5 Instant: 52.0 (#94)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Non-English | 1439 | 1406 |
| LMArena Chinese | 1486 | 1449 |
| LMArena French | 1472 | 1403 |
| LMArena German | 1458 | 1413 |
| LMArena Korean | 1447 | 1378 |
| LMArena Russian | 1453 | 1404 |
| LMArena Spanish | 1458 | 1447 |
| LMArena Japanese | 1445 | — |
Instruction Following Too close to call
DeepSeek V4 Pro: 76.1 (#47), Kimi K2.5 Instant: 75.3 (#65)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Instruction Following | 1448 | 1430 |
Long Context DeepSeek V4 Pro leads
DeepSeek V4 Pro: 45.0 (#51), Kimi K2.5 Instant: 43.9 (#83)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Longer Query | 1458 | 1435 |
| CL-bench Life | 13.5% | — |
Writing & Preference DeepSeek V4 Pro leads
DeepSeek V4 Pro: 65.5 (#46), Kimi K2.5 Instant: 60.6 (#95)
| Benchmark | DeepSeek V4 Pro | Kimi K2.5 Instant |
|---|---|---|
| LMArena Text | 1451 | 1420 |
| LMArena Creative Writing | 1446 | 1381 |
| LMArena Multi-Turn | 1467 | 1427 |
| EQ-Bench Creative Writing | 1553 | — |
| EQ-Bench 4 | 1166 | — |
Frequently asked questions
Is DeepSeek V4 Pro better than Kimi K2.5 Instant?
DeepSeek V4 Pro is the stronger model overall, scoring 54.3 to 43.6 on the Noometry Index.
Is DeepSeek V4 Pro or Kimi K2.5 Instant better for coding?
DeepSeek V4 Pro scores higher on coding benchmarks: 52.4 versus 42.6 in the Noometry coding category.
How many benchmarks do DeepSeek V4 Pro and Kimi K2.5 Instant share?
17 benchmarks have published results for both models. DeepSeek V4 Pro has 48 scored results on Noometry and Kimi K2.5 Instant has 18.