Model comparison
Kimi K2.5 vs Llama 3.1-405B
Kimi K2.5 is the stronger model overall, scoring 48.1 to 30.7 on the Noometry Index.
Last verified . 24 shared benchmarks.
MetaSummary
- They share 24 benchmarks with published results for both. Kimi K2.5 scores higher in 9 categories and Llama 3.1-405B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2.5 leads 51.8 to 18.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 92.2% for Kimi K2.5 and 9.7% for Llama 3.1-405B.
Side by side
| Kimi K2.5 | Llama 3.1-405B | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 48.1 | 30.7 |
| Released | 2026-01-27 | 2024-07-23 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 262K | — |
| Input $ / M tokens | $0.45 | — |
| Output $ / M tokens | $2.25 | — |
| Results tracked | 51 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Kimi K2.5: 48.8 (#53), Llama 3.1-405B: 33.1 (#262)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| WeirdML | 45.6% | 21.4% |
| LMArena Coding | 1474 | 1291 |
| SWE-bench Verified | 73.8% | — |
| SWE-bench Verified (bash only) | 70.8% | — |
| LMArena WebDev | 1437 | — |
| SWE-bench Multilingual | 67.3% | — |
| SciCode | 49% | — |
| ALE-Bench | 821.65 | — |
Agentic & Tool Use Kimi K2.5 leads
Kimi K2.5: 34.2 (#48), Llama 3.1-405B: 21.0 (#140)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| Terminal-Bench | 43.2% | — |
| TheAgentCompany | — | 7.4% |
| Cybench | — | 7.5% |
| OSWorld | 63.3% | — |
| Vending-Bench 2 | 1,198 | — |
Reasoning Kimi K2.5 leads
Kimi K2.5: 31.2 (#80), Llama 3.1-405B: 16.8 (#300)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| SimpleBench | 46.8% | 23% |
| Kagi LLM Benchmark | 78.5% | 45% |
| LMArena Hard Prompts | 1453 | 1269 |
| Epoch Capabilities Index | 148.03 | 128.75 |
| ARC-AGI-2 | 11.8% | — |
| NYT Connections (extended) | 69.9% | — |
| ARC-AGI-1 | 65.3% | — |
| CritPt | 3.1% | — |
| Chess Puzzles | 12% | — |
| EnigmaEval | 3.4% | — |
| Thematic Generalization | 69.4% | — |
| DTBench | — | 61.4% |
| BIG-Bench Hard | — | 82.9% |
| ForecastBench | — | 59.9 |
| HellaSwag | — | 89.2% |
| PIQA | — | 85.9% |
| WinoGrande | — | 89.2% |
Math Kimi K2.5 leads
Kimi K2.5: 51.8 (#53), Llama 3.1-405B: 18.4 (#290)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 92.2% | 9.7% |
| LMArena Math | 1470 | 1281 |
| MathArena Final-Answer Competitions | 62.3% | — |
| Omni-MATH | — | 24.9% |
| MATH Level 5 | — | 49.8% |
| FrontierMath (Feb 2025 set) | 27.9% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Kimi K2.5 leads
Kimi K2.5: 53.6 (#56), Llama 3.1-405B: 30.4 (#227)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| GPQA Diamond | 87.6% | 50.9% |
| LMArena Expert | 1466 | 1243 |
| Humanity's Last Exam | 24.4% | — |
| SimpleQA Verified | 34.3% | — |
| MMLU-Pro | — | 72.3% |
| Confabulations | — | 17.6% |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | — | 52.2% |
| ARC (AI2) Challenge | — | 95.3% |
| MMLU | — | 84.5% |
| TriviaQA | — | 82.7% |
Multimodal Not comparable
Kimi K2.5: 41.1 (#39), Llama 3.1-405B: —
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| LMArena Vision | 1269 | — |
| LMArena Document | 1430 | — |
Multilingual Kimi K2.5 leads
Kimi K2.5: 53.9 (#53), Llama 3.1-405B: 40.7 (#214)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| LMArena Non-English | 1433 | 1248 |
| LMArena Chinese | 1495 | 1242 |
| LMArena French | 1454 | 1279 |
| LMArena German | 1441 | 1252 |
| LMArena Japanese | 1421 | 1208 |
| LMArena Korean | 1410 | 1184 |
| LMArena Russian | 1435 | 1265 |
| LMArena Spanish | 1450 | 1260 |
Instruction Following Kimi K2.5 leads
Kimi K2.5: 75.3 (#64), Llama 3.1-405B: 65.9 (#214)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| LMArena Instruction Following | 1431 | 1259 |
| IFEval | — | 81.1% |
Long Context Kimi K2.5 leads
Kimi K2.5: 52.1 (#7), Llama 3.1-405B: 38.4 (#197)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| LMArena Longer Query | 1445 | 1266 |
| Fiction.LiveBench | 86.1% | — |
| CL-bench | 19.3% | — |
| CL-bench Life | 13.2% | — |
Writing & Preference Kimi K2.5 leads
Kimi K2.5: 65.1 (#53), Llama 3.1-405B: 38.9 (#251)
| Benchmark | Kimi K2.5 | Llama 3.1-405B |
|---|---|---|
| LMArena Text | 1445 | 1284 |
| LMArena Creative Writing | 1423 | 1262 |
| EQ-Bench Creative Writing | 1579 | 870 |
| LMArena Multi-Turn | 1444 | 1297 |
| WildBench | — | 78.3% |
Frequently asked questions
Is Kimi K2.5 better than Llama 3.1-405B?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 30.7 on the Noometry Index.
Is Kimi K2.5 or Llama 3.1-405B better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 33.1 in the Noometry coding category.
How many benchmarks do Kimi K2.5 and Llama 3.1-405B share?
24 benchmarks have published results for both models. Kimi K2.5 has 51 scored results on Noometry and Llama 3.1-405B has 42.