Model comparison
Kimi K2 Thinking Turbo vs Llama-3.3-70B-Instruct
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 30.6 on the Noometry Index.
Last verified . 19 shared benchmarks.
MetaSummary
- They share 19 benchmarks with published results for both. Kimi K2 Thinking Turbo scores higher in 8 categories and Llama-3.3-70B-Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K2 Thinking Turbo leads 47.4 to 15.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 83.1% for Kimi K2 Thinking Turbo and 5.1% for Llama-3.3-70B-Instruct.
Side by side
| Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 45.8 | 30.6 |
| Released | 2025-11-06 | 2024-12-06 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.32 |
| Results tracked | 21 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 38.4 (#178), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Coding | 1454 | 1268 |
| LMArena WebDev | 1322 | — |
| SciCode | — | 26% |
| WeirdML | — | 14.4% |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
Agentic & Tool Use Not comparable
Kimi K2 Thinking Turbo: —, Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |
Reasoning Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 32.5 (#75), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Hard Prompts | 1428 | 1257 |
| SimpleBench | — | 19.9% |
| CritPt | — | 0% |
| Chess Puzzles | 20% | — |
| LiveBench Reasoning | — | 50.8% |
| DTBench | — | 59.5% |
| LiveBench Data Analysis | — | 49.5% |
| LMCA | — | 17.5% |
| Epoch Capabilities Index | — | 127.33 |
| ForecastBench | — | 58.6 |
| LiveBench | — | 50.2% |
Math Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 47.4 (#68), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 83.1% | 5.1% |
| LMArena Math | 1429 | 1267 |
| LiveBench Math | — | 42.2% |
| MATH Level 5 | — | 41.6% |
Knowledge Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 50.9 (#69), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 84.2% | 47.4% |
| LMArena Expert | 1439 | 1225 |
| Confabulations | — | 22.8% |
| Vectara Hallucination Rate | — | 4.1% |
| MMLU | — | 86.3% |
Multilingual Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 51.4 (#109), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1398 | 1236 |
| LMArena Chinese | 1456 | 1217 |
| LMArena French | 1425 | 1281 |
| LMArena German | 1390 | 1251 |
| LMArena Japanese | 1357 | 1150 |
| LMArena Korean | 1330 | 1143 |
| LMArena Russian | 1391 | 1252 |
| LMArena Spanish | 1406 | 1270 |
Instruction Following Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 74.0 (#109), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1403 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 43.2 (#102), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1415 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Kimi K2 Thinking Turbo leads
Kimi K2 Thinking Turbo: 60.0 (#104), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | Kimi K2 Thinking Turbo | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1415 | 1274 |
| LMArena Creative Writing | 1374 | 1250 |
| LMArena Multi-Turn | 1414 | 1280 |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is Kimi K2 Thinking Turbo better than Llama-3.3-70B-Instruct?
Kimi K2 Thinking Turbo is the stronger model overall, scoring 45.8 to 30.6 on the Noometry Index.
Is Kimi K2 Thinking Turbo or Llama-3.3-70B-Instruct better for coding?
Kimi K2 Thinking Turbo scores higher on coding benchmarks: 38.4 versus 31.0 in the Noometry coding category.
How many benchmarks do Kimi K2 Thinking Turbo and Llama-3.3-70B-Instruct share?
19 benchmarks have published results for both models. Kimi K2 Thinking Turbo has 21 scored results on Noometry and Llama-3.3-70B-Instruct has 43.