Model comparison
Gemma 4 31B IT vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.5 to 43.2 on the Noometry Index. Gemma 4 31B IT costs 11× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Last verified . 32 shared benchmarks.
GoogleSummary
- They share 32 benchmarks with published results for both. Gemma 4 31B IT scores higher in 1 category and Kimi K2.6 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Kimi K2.6 leads 53.9 to 37.9.
- The biggest single-benchmark swing is SimpleQA Verified: 10.4% for Gemma 4 31B IT and 34.9% for Kimi K2.6.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
Side by side
| Gemma 4 31B IT | Kimi K2.6 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 43.2 | 47.5 |
| Released | 2026-04-02 | 2026-04-20 |
| Weights | Open | Open |
| Context window | 262K | 262K |
| Max output | 33K | 262K |
| Input $ / M tokens | $0.09 | $0.95 |
| Output $ / M tokens | $0.34 | $4 |
| Results tracked | 35 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Gemma 4 31B IT: 41.9 (#98), Kimi K2.6: 50.5 (#41)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| LMArena WebDev | 1364 | 1509 |
| SciCode | 43.4% | 53.5% |
| WeirdML | 52.3% | 55.9% |
| LMArena Coding | 1459 | 1488 |
| ALE-Bench | 925.5 | 1,093 |
| SWE-bench Verified | — | 76.7% |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Kimi K2.6: 21.8 (#128)
| Benchmark | Gemma 4 31B IT | 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
Gemma 4 31B IT: 26.4 (#119), Kimi K2.6: 39.6 (#51)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| NYT Connections (extended) | 70.6% | 87.2% |
| CritPt | 1.4% | 8% |
| Chess Puzzles | 5% | 26% |
| LMArena Hard Prompts | 1448 | 1470 |
| DTBench | 82.7% | 90.9% |
| LMCA | 39.3% | 37.3% |
| Epoch Capabilities Index | 142.74 | 151.03 |
| Kagi LLM Benchmark | 63.5% | — |
| Thematic Generalization | 53% | — |
| EBR-Bench | — | 2.4% |
| Mystery Game Puzzles | — | 18% |
| Surface Evolver Bench | 30.6% | — |
Math Kimi K2.6 leads
Gemma 4 31B IT: 41.3 (#72), Kimi K2.6: 56.8 (#34)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 96.1% |
| LMArena Math | 1464 | 1475 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Kimi K2.6 leads
Gemma 4 31B IT: 37.9 (#141), Kimi K2.6: 53.9 (#48)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 75.8% | 90.8% |
| SimpleQA Verified | 10.4% | 34.9% |
| Vectara Hallucination Rate | 7.4% | 10.8% |
| LMArena Expert | 1465 | 1491 |
Multimodal Gemma 4 31B IT leads
Gemma 4 31B IT: 41.0 (#32), Kimi K2.6: 31.8 (#95)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1278 | 1282 |
| LMArena Document | 1425 | 1451 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
Multilingual Kimi K2.6 leads
Gemma 4 31B IT: 54.0 (#52), Kimi K2.6: 55.1 (#35)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1431 | 1446 |
| LMArena Chinese | 1478 | 1521 |
| LMArena French | 1436 | 1471 |
| LMArena Russian | 1462 | 1446 |
| LMArena Spanish | 1443 | 1463 |
| LMArena German | — | 1449 |
| LMArena Japanese | — | 1442 |
| LMArena Korean | — | 1428 |
Instruction Following Too close to call
Gemma 4 31B IT: 75.6 (#58), Kimi K2.6: 76.5 (#40)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1451 |
Long Context Too close to call
Gemma 4 31B IT: 44.2 (#67), Kimi K2.6: 44.9 (#50)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1446 | 1468 |
Writing & Preference Kimi K2.6 leads
Gemma 4 31B IT: 60.7 (#92), Kimi K2.6: 68.7 (#25)
| Benchmark | Gemma 4 31B IT | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1443 | 1455 |
| LMArena Creative Writing | 1416 | 1434 |
| EQ-Bench Creative Writing | 1368 | 1725 |
| EQ-Bench 4 | 1120 | 1202 |
| LMArena Multi-Turn | 1452 | 1453 |
Frequently asked questions
Is Gemma 4 31B IT better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.5 to 43.2 on the Noometry Index. Gemma 4 31B IT costs 11× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Which is cheaper, Gemma 4 31B IT or Kimi K2.6?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is Gemma 4 31B IT or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 41.9 in the Noometry coding category.
Which has the bigger context window?
Both accept 262K tokens.
How many benchmarks do Gemma 4 31B IT and Kimi K2.6 share?
32 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Kimi K2.6 has 51.