Model comparison
Kimi K2.6 vs Llama 3-70B
Kimi K2.6 is the stronger model overall, scoring 47.7 to 28.8 on the Noometry Index.
Last verified . 21 shared benchmarks.
MetaSummary
- They share 21 benchmarks with published results for both. Kimi K2.6 scores higher in 9 categories and Llama 3-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.6 leads 57.0 to 12.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 96.1% for Kimi K2.6 and 4.3% for Llama 3-70B.
Side by side
| Kimi K2.6 | Llama 3-70B | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 47.7 | 28.8 |
| Released | 2026-04-20 | 2024-04-18 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 262K | — |
| Input $ / M tokens | $0.95 | — |
| Output $ / M tokens | $4 | — |
| Results tracked | 51 | 31 |
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 3-70B: 35.8 (#218)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Coding | 1488 | 1206 |
| SWE-bench Verified | 76.7% | — |
| LMArena WebDev | 1509 | — |
| SciCode | 53.5% | — |
| WeirdML | 55.9% | — |
| BigCodeBench Instruct | — | 43.6% |
| BigCodeBench Complete | — | 54.5% |
| ALE-Bench | 1,093 | — |
| HumanEval+ | — | 72% |
| MBPP+ | — | 69% |
Agentic & Tool Use Too close to call
Kimi K2.6: 21.9 (#137), Llama 3-70B: 21.1 (#139)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| OSWorld 2.0 | 4.6% | — |
| Cybench | — | 5% |
| 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 3-70B: 18.0 (#288)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Hard Prompts | 1470 | 1195 |
| DTBench | 90.9% | 54.2% |
| Epoch Capabilities Index | 151.05 | 122.93 |
| Kagi LLM Benchmark | — | 35.1% |
| NYT Connections (extended) | 87.2% | — |
| CritPt | 8% | — |
| Chess Puzzles | 26% | — |
| EBR-Bench | 2.4% | — |
| Mystery Game Puzzles | 18% | — |
| LMCA | 37.3% | — |
| ForecastBench | — | 57.1 |
| WinoGrande | — | 83.5% |
Math Kimi K2.6 leads
Kimi K2.6: 57.0 (#41), Llama 3-70B: 12.8 (#305)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 96.1% | 4.3% |
| LMArena Math | 1475 | 1218 |
| FrontierMath (Tiers 1-3) | 57.2% | — |
| FrontierMath Tier 4 | 25.6% | — |
| MathArena Final-Answer Competitions | 72.9% | — |
| ProofBench | 16% | — |
| MATH Level 5 | — | 22.6% |
| FrontierMath (Feb 2025 set) | 39% | — |
| FrontierMath Tier 4 (v1) | 14.6% | — |
Knowledge Kimi K2.6 leads
Kimi K2.6: 54.0 (#54), Llama 3-70B: 20.8 (#277)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| GPQA Diamond | 90.8% | 40.6% |
| LMArena Expert | 1491 | 1149 |
| SimpleQA Verified | 34.9% | — |
| Vectara Hallucination Rate | 10.8% | — |
| MMLU | — | 79.3% |
Multimodal Not comparable
Kimi K2.6: 31.6 (#103), Llama 3-70B: —
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Vision | 1283 | — |
| Blueprint-Bench 2 | 3.9% | — |
| Furniture Assembly | 21.7% | — |
| LMArena Document | 1451 | — |
Multilingual Kimi K2.6 leads
Kimi K2.6: 54.9 (#37), Llama 3-70B: 33.6 (#251)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Non-English | 1446 | 1142 |
| LMArena Chinese | 1521 | 1114 |
| LMArena French | 1471 | 1232 |
| LMArena German | 1450 | 1169 |
| LMArena Japanese | 1443 | 1017 |
| LMArena Korean | 1427 | 1017 |
| LMArena Russian | 1446 | 1159 |
| LMArena Spanish | 1464 | 1241 |
Instruction Following Kimi K2.6 leads
Kimi K2.6: 76.3 (#43), Llama 3-70B: 62.5 (#238)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Instruction Following | 1451 | 1194 |
Long Context Kimi K2.6 leads
Kimi K2.6: 44.9 (#52), Llama 3-70B: 35.6 (#240)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Longer Query | 1468 | 1174 |
Writing & Preference Kimi K2.6 leads
Kimi K2.6: 68.5 (#26), Llama 3-70B: 42.8 (#231)
| Benchmark | Kimi K2.6 | Llama 3-70B |
|---|---|---|
| LMArena Text | 1455 | 1221 |
| LMArena Creative Writing | 1434 | 1210 |
| LMArena Multi-Turn | 1453 | 1223 |
| EQ-Bench Creative Writing | 1725 | — |
| EQ-Bench 4 | 1202 | — |
Frequently asked questions
Is Kimi K2.6 better than Llama 3-70B?
Kimi K2.6 is the stronger model overall, scoring 47.7 to 28.8 on the Noometry Index.
Is Kimi K2.6 or Llama 3-70B better for coding?
Kimi K2.6 scores higher on coding benchmarks: 50.7 versus 35.8 in the Noometry coding category.
How many benchmarks do Kimi K2.6 and Llama 3-70B share?
21 benchmarks have published results for both models. Kimi K2.6 has 51 scored results on Noometry and Llama 3-70B has 31.