Model comparison
Kimi K2.6 vs Llama 4 Maverick
Kimi K2.6 is the stronger model overall, scoring 47.7 to 30.9 on the Noometry Index. Llama 4 Maverick costs 5.6× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Last verified . 31 shared benchmarks.
MetaSummary
- They share 31 benchmarks with published results for both. Kimi K2.6 scores higher in 9 categories and Llama 4 Maverick in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 57.0 to 26.0.
- The biggest single-benchmark swing is NYT Connections (extended): 87.2% for Kimi K2.6 and 8% for Llama 4 Maverick.
- Llama 4 Maverick is cheaper at $0.19 / $0.65 per million input/output tokens, against $0.95 / $4 for Kimi K2.6.
- Kimi K2.6 accepts more context: 262K tokens versus 128K.
Side by side
| Kimi K2.6 | Llama 4 Maverick | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 47.7 | 30.9 |
| Released | 2026-04-20 | 2025-04-05 |
| Weights | Open | Open |
| Context window | 262K | 128K |
| Max output | 262K | 4K |
| Input $ / M tokens | $0.95 | $0.19 |
| Output $ / M tokens | $4 | $0.65 |
| Results tracked | 51 | 54 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.6 leads
Kimi K2.6: 50.7 (#43), Llama 4 Maverick: 26.6 (#324)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| SciCode | 53.5% | 33.1% |
| WeirdML | 55.9% | 24.5% |
| LMArena Coding | 1488 | 1302 |
| ALE-Bench | 1,093 | 172.97 |
| SWE-bench Verified | 76.7% | — |
| SWE-bench Verified (bash only) | — | 21% |
| Aider Polyglot | — | 15.6% |
| LMArena WebDev | 1509 | — |
| BigCodeBench Instruct | — | 49.7% |
| BigCodeBench Complete | — | 61.4% |
Agentic & Tool Use Llama 4 Maverick leads
Kimi K2.6: 21.9 (#137), Llama 4 Maverick: 28.2 (#91)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 37.3% |
| OSWorld 2.0 | 4.6% | — |
| ExploitBench | 18.4% | — |
| GBAEval | 0.9% | — |
| GDP.pdf | 12% | — |
| Vending-Bench 2 | 6,205 | — |
Reasoning Kimi K2.6 leads
Kimi K2.6: 40.5 (#55), Llama 4 Maverick: 10.1 (#342)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| NYT Connections (extended) | 87.2% | 8% |
| CritPt | 8% | 0% |
| LMArena Hard Prompts | 1470 | 1281 |
| DTBench | 90.9% | 61.9% |
| LMCA | 37.3% | 15.9% |
| Epoch Capabilities Index | 151.05 | 132.2 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 27.7% |
| Kagi LLM Benchmark | — | 55.9% |
| ARC-AGI-1 | — | 4.4% |
| Chess Puzzles | 26% | — |
| EnigmaEval | — | 0.6% |
| EBR-Bench | 2.4% | — |
| Mystery Game Puzzles | 18% | — |
| ForecastBench | — | 57.5 |
Math Kimi K2.6 leads
Kimi K2.6: 57.0 (#41), Llama 4 Maverick: 26.0 (#262)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 96.1% | 20.6% |
| LMArena Math | 1475 | 1299 |
| FrontierMath (Feb 2025 set) | 39% | 0.7% |
| FrontierMath (Tiers 1-3) | 57.2% | — |
| FrontierMath Tier 4 | 25.6% | — |
| MathArena Final-Answer Competitions | 72.9% | — |
| ProofBench | 16% | — |
| Omni-MATH | — | 42.2% |
| MATH Level 5 | — | 73% |
| FrontierMath Tier 4 (v1) | 14.6% | — |
Knowledge Kimi K2.6 leads
Kimi K2.6: 54.0 (#54), Llama 4 Maverick: 33.4 (#204)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| GPQA Diamond | 90.8% | 67% |
| Vectara Hallucination Rate | 10.8% | 8.2% |
| LMArena Expert | 1491 | 1259 |
| Humanity's Last Exam | — | 5.7% |
| SimpleQA Verified | 34.9% | — |
| MMLU-Pro | — | 81% |
| Confabulations | — | 22.6% |
| GPQA (HELM) | — | 65% |
Multimodal Too close to call
Kimi K2.6: 31.6 (#103), Llama 4 Maverick: 31.6 (#105)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| LMArena Vision | 1283 | 1142 |
| GeoBench | — | 52% |
| Blueprint-Bench 2 | 3.9% | — |
| Furniture Assembly | 21.7% | — |
| LMArena Document | 1451 | — |
| SpatialViz-Bench | — | 31.8% |
Multilingual Kimi K2.6 leads
Kimi K2.6: 54.9 (#37), Llama 4 Maverick: 42.2 (#195)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| LMArena Non-English | 1446 | 1269 |
| LMArena Chinese | 1521 | 1277 |
| LMArena French | 1471 | 1259 |
| LMArena German | 1450 | 1291 |
| LMArena Japanese | 1443 | 1207 |
| LMArena Korean | 1427 | 1203 |
| LMArena Russian | 1446 | 1286 |
| LMArena Spanish | 1464 | 1293 |
Instruction Following Kimi K2.6 leads
Kimi K2.6: 76.3 (#43), Llama 4 Maverick: 71.7 (#146)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| LMArena Instruction Following | 1451 | 1267 |
| IFEval | — | 90.8% |
Long Context Kimi K2.6 leads
Kimi K2.6: 44.9 (#52), Llama 4 Maverick: 31.4 (#279)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| LMArena Longer Query | 1468 | 1280 |
| Fiction.LiveBench | — | 46.2% |
Writing & Preference Kimi K2.6 leads
Kimi K2.6: 68.5 (#26), Llama 4 Maverick: 38.8 (#252)
| Benchmark | Kimi K2.6 | Llama 4 Maverick |
|---|---|---|
| LMArena Text | 1455 | 1287 |
| LMArena Creative Writing | 1434 | 1267 |
| EQ-Bench Creative Writing | 1725 | 860 |
| LMArena Multi-Turn | 1453 | 1289 |
| Short-Story Creative Writing | — | 62% |
| WildBench | — | 80% |
| EQ-Bench 4 | 1202 | — |
Frequently asked questions
Is Kimi K2.6 better than Llama 4 Maverick?
Kimi K2.6 is the stronger model overall, scoring 47.7 to 30.9 on the Noometry Index. Llama 4 Maverick costs 5.6× less per token, which makes it the better buy when Kimi K2.6's lead doesn't matter for your workload.
Which is cheaper, Kimi K2.6 or Llama 4 Maverick?
Llama 4 Maverick is cheaper. It lists at $0.19 per million input tokens and $0.65 per million output tokens; Kimi K2.6 lists at $0.95 and $4.
Is Kimi K2.6 or Llama 4 Maverick better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 26.6 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.6 does, with 262K tokens against 128K.
How many benchmarks do Kimi K2.6 and Llama 4 Maverick share?
31 benchmarks have published results for both models. Kimi K2.6 has 51 scored results on Noometry and Llama 4 Maverick has 54.