Model comparison
Gemma 2 27B vs Kimi K2.6
Kimi K2.6 is the stronger model overall, scoring 47.5 to 29.2 on the Noometry Index. Gemma 2 27B costs 2.6× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
GoogleSummary
- They share 22 benchmarks with published results for both. Gemma 2 27B scores higher in 0 categories and Kimi K2.6 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 56.8 to 10.0.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.4% for Gemma 2 27B and 96.1% for Kimi K2.6.
- Gemma 2 27B is cheaper at $0.65 / $0.65 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Kimi K2.6 accepts more context: 262K tokens versus 8K.
Side by side
| Gemma 2 27B | Kimi K2.6 | |
|---|---|---|
| Provider | Moonshot AI | |
| Noometry Index | 29.2 | 47.5 |
| Released | 2024-06-24 | 2026-04-20 |
| Weights | Open | Open |
| Context window | 8K | 262K |
| Max output | 2K | 262K |
| Input $ / M tokens | $0.65 | $0.95 |
| Output $ / M tokens | $0.65 | $4 |
| Results tracked | 34 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Gemma 2 27B: 33.5 (#243), Kimi K2.6: 50.5 (#41)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Coding | 1211 | 1488 |
| SWE-bench Verified | — | 76.7% |
| LMArena WebDev | — | 1509 |
| SciCode | — | 53.5% |
| WeirdML | — | 55.9% |
| BigCodeBench Instruct | 42.8% | — |
| LiveBench Coding | 36% | — |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | — | 1,093 |
Agentic & Tool Use Not comparable
Gemma 2 27B: —, Kimi K2.6: 21.8 (#128)
| Benchmark | Gemma 2 27B | 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 2 27B: 14.8 (#313), Kimi K2.6: 39.6 (#51)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Hard Prompts | 1199 | 1470 |
| DTBench | 48% | 90.9% |
| LMCA | 7.1% | 37.3% |
| Epoch Capabilities Index | 122.06 | 151.03 |
| NYT Connections (extended) | — | 87.2% |
| CritPt | — | 8% |
| Chess Puzzles | — | 26% |
| EBR-Bench | — | 2.4% |
| LiveBench Reasoning | 28.1% | — |
| Mystery Game Puzzles | — | 18% |
| LiveBench Data Analysis | 47.9% | — |
| LiveBench | 38.2% | — |
Math Kimi K2.6 leads
Gemma 2 27B: 10.0 (#306), Kimi K2.6: 56.8 (#34)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 96.1% |
| LMArena Math | 1212 | 1475 |
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 25.6% |
| MathArena Final-Answer Competitions | — | 72.9% |
| ProofBench | — | 16% |
| LiveBench Math | 26.5% | — |
| MATH Level 5 | 27.9% | — |
| FrontierMath (Feb 2025 set) | — | 39% |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge Kimi K2.6 leads
Gemma 2 27B: 18.9 (#275), Kimi K2.6: 53.9 (#48)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 36.5% | 90.8% |
| LMArena Expert | 1172 | 1491 |
| SimpleQA Verified | — | 34.9% |
| Confabulations | 27.1% | — |
| Vectara Hallucination Rate | — | 10.8% |
| MMLU | 75.7% | — |
Multimodal Not comparable
Gemma 2 27B: —, Kimi K2.6: 31.8 (#95)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Vision | — | 1282 |
| Blueprint-Bench 2 | — | 3.9% |
| Furniture Assembly | — | 21.7% |
| LMArena Document | — | 1451 |
Multilingual Kimi K2.6 leads
Gemma 2 27B: 38.8 (#221), Kimi K2.6: 55.1 (#35)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Non-English | 1217 | 1446 |
| LMArena Chinese | 1221 | 1521 |
| LMArena French | 1246 | 1471 |
| LMArena German | 1209 | 1449 |
| LMArena Japanese | 1175 | 1442 |
| LMArena Korean | 1174 | 1428 |
| LMArena Russian | 1234 | 1446 |
| LMArena Spanish | 1228 | 1463 |
Instruction Following Kimi K2.6 leads
Gemma 2 27B: 60.7 (#244), Kimi K2.6: 76.5 (#40)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Instruction Following | 1206 | 1451 |
| LiveBench Instruction Following | 58.1% | — |
Long Context Kimi K2.6 leads
Gemma 2 27B: 37.3 (#213), Kimi K2.6: 44.9 (#50)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Longer Query | 1231 | 1468 |
Writing & Preference Kimi K2.6 leads
Gemma 2 27B: 44.4 (#220), Kimi K2.6: 68.7 (#25)
| Benchmark | Gemma 2 27B | Kimi K2.6 |
|---|---|---|
| LMArena Text | 1231 | 1455 |
| LMArena Creative Writing | 1241 | 1434 |
| LMArena Multi-Turn | 1224 | 1453 |
| EQ-Bench Creative Writing | — | 1725 |
| EQ-Bench 4 | — | 1202 |
| LiveBench Language | 32.6% | — |
Frequently asked questions
Is Gemma 2 27B better than Kimi K2.6?
Kimi K2.6 is the stronger model overall, scoring 47.5 to 29.2 on the Noometry Index. Gemma 2 27B costs 2.6× 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 2 27B or Kimi K2.6?
Gemma 2 27B is cheaper. It lists at $0.65 per million input tokens and $0.65 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is Gemma 2 27B or Kimi K2.6 better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.5 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 8K.
How many benchmarks do Gemma 2 27B and Kimi K2.6 share?
22 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Kimi K2.6 has 51.