Model comparison
Gemma 3 27B vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.7 to 30.8 on the Noometry Index. Gemma 3 27B costs 17× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Last verified . 28 shared benchmarks.
GoogleSummary
- They share 28 benchmarks with published results for both. Gemma 3 27B scores higher in 2 categories and Kimi K2.6 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 57.0 to 25.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 22.5% for Gemma 3 27B and 96.1% for Kimi K2.6.
- Gemma 3 27B is cheaper at $0.08 / $0.16 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Kimi K2.6 accepts more context: 262K tokens versus 131K.
Side by side
| Gemma 3 27B | Kimi K2.6 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 30.8 | 47.7 |
| Released | 2025-03-11 | 2026-04-20 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $0.08 | $0.95 |
| Output $ / M tokens | $0.16 | $4 |
| Results tracked | 43 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Gemma 3 27B: 22.5 (#334), Kimi K2.6: 50.7 (#43)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| SciCode | 21.2% | 53.5% |
| LMArena Coding | 1322 | 1488 |
| SWE-bench Verified | — | 76.7% |
| Aider Polyglot | 4.9% | — |
| LMArena WebDev | — | 1509 |
| WeirdML | — | 55.9% |
| LiveBench Coding | 39.9% | — |
| ALE-Bench | — | 1,093 |
Agentic & Tool Use Gemma 3 27B leads
Gemma 3 27B: 25.1 (#110), Kimi K2.6: 21.9 (#137)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | — |
| 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 3 27B: 16.7 (#301), Kimi K2.6: 40.5 (#55)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| CritPt | 0% | 8% |
| Chess Puzzles | 0% | 26% |
| LMArena Hard Prompts | 1340 | 1470 |
| DTBench | 52.5% | 90.9% |
| LMCA | 12.3% | 37.3% |
| Epoch Capabilities Index | 130.04 | 151.05 |
| Kagi LLM Benchmark | 40.4% | — |
| NYT Connections (extended) | — | 87.2% |
| EBR-Bench | — | 2.4% |
| LiveBench Reasoning | 43.8% | — |
| Mystery Game Puzzles | — | 18% |
| LiveBench Data Analysis | 51.5% | — |
| LiveBench | 50% | — |
Math Kimi K2.6 leads
Gemma 3 27B: 25.9 (#265), Kimi K2.6: 57.0 (#41)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 22.5% | 96.1% |
| LMArena Math | 1312 | 1475 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| LiveBench Math | 55.4% | — |
| MATH Level 5 | 74% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Kimi K2.6 leads
Gemma 3 27B: 25.5 (#261), Kimi K2.6: 54.0 (#54)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 47.7% | 90.8% |
| Vectara Hallucination Rate | 7.4% | 10.8% |
| LMArena Expert | 1304 | 1491 |
| SimpleQA Verified | — | 34.9% |
| Confabulations | 40.3% | — |
Multimodal Too close to call
Gemma 3 27B: 32.6 (#100), Kimi K2.6: 31.6 (#103)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| LMArena Vision | 1164 | 1283 |
| GeoBench | 52% | — |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Kimi K2.6 leads
Gemma 3 27B: 46.9 (#155), Kimi K2.6: 54.9 (#37)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1334 | 1446 |
| LMArena Chinese | 1346 | 1521 |
| LMArena French | 1368 | 1471 |
| LMArena German | 1362 | 1450 |
| LMArena Japanese | 1287 | 1443 |
| LMArena Korean | 1308 | 1427 |
| LMArena Russian | 1349 | 1446 |
| LMArena Spanish | 1349 | 1464 |
Instruction Following Kimi K2.6 leads
Gemma 3 27B: 70.6 (#160), Kimi K2.6: 76.3 (#43)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1321 | 1451 |
| LiveBench Instruction Following | 74.9% | — |
Long Context Kimi K2.6 leads
Gemma 3 27B: 27.6 (#293), Kimi K2.6: 44.9 (#52)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1333 | 1468 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Kimi K2.6 leads
Gemma 3 27B: 52.5 (#168), Kimi K2.6: 68.5 (#26)
| Benchmark | Gemma 3 27B | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1358 | 1455 |
| LMArena Creative Writing | 1346 | 1434 |
| EQ-Bench Creative Writing | 1266 | 1725 |
| LMArena Multi-Turn | 1345 | 1453 |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench 4 | — | 1202 |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.7 to 30.8 on the Noometry Index. Gemma 3 27B costs 17× 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 3 27B or Kimi K2.6?
Gemma 3 27B is cheaper. It lists at $0.08 per million input tokens and $0.16 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is Gemma 3 27B or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 22.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 131K.
How many benchmarks do Gemma 3 27B and Kimi K2.6 share?
28 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Kimi K2.6 has 51.