Model comparison
Kimi K2 (Jul 2025) vs Sonar
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.5 on the Noometry Index.
Last verified . 0 shared benchmarks.
PerplexitySummary
- The widest gap is in writing & preference, where Kimi K2 (Jul 2025) leads 62.3 to 52.6.
- Both cost about the same: $0.57 input and $2.30 output per million tokens.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 128K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Kimi K2 (Jul 2025) | Sonar | |
|---|---|---|
| Provider | Moonshot AI | Perplexity |
| Noometry Index | 41.2 | 38.5 |
| Released | 2025-07-12 | 2024-01-01 |
| Weights | Open | Proprietary |
| Context window | 262K | 128K |
| Max output | 262K | 4K |
| Input $ / M tokens | $0.57 | $1 |
| Output $ / M tokens | $2.30 | $1 |
| Results tracked | 42 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.4 (#102), Sonar: 35.7 (#221)
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| GSO | 4.9% | — |
| WeirdML | 42.8% | — |
| LiveBench Coding | — | 35.1% |
| LMArena Coding | 1399 | — |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Not comparable
Kimi K2 (Jul 2025): 32.4 (#64), Sonar: —
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| METR Time Horizons | 59.2% | — |
Reasoning Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 23.3 (#179), Sonar: 21.1 (#227)
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| SimpleBench | 26.3% | — |
| Kagi LLM Benchmark | 64.4% | — |
| LiveBench Reasoning | — | 46.3% |
| LMArena Hard Prompts | 1384 | — |
| LiveBench Data Analysis | — | 37.9% |
| Epoch Capabilities Index | 146.01 | — |
| ForecastBench | 60.2 | — |
| LiveBench | — | 46.9% |
Math Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 42.7 (#83), Sonar: 33.7 (#200)
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| Omni-MATH | 65.4% | — |
| LiveBench Math | — | 41.6% |
| LMArena Math | 1397 | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Not comparable
Kimi K2 (Jul 2025): 37.3 (#157), Sonar: —
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| MMLU-Pro | 81.9% | — |
| Confabulations | 20.4% | — |
| Vectara Hallucination Rate | 17.9% | — |
| GPQA (HELM) | 65.3% | — |
| LMArena Expert | 1365 | — |
Multilingual Not comparable
Kimi K2 (Jul 2025): 49.6 (#130), Sonar: —
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| LMArena Non-English | 1372 | — |
| LMArena Chinese | 1415 | — |
| LMArena French | 1379 | — |
| LMArena German | 1387 | — |
| LMArena Japanese | 1349 | — |
| LMArena Korean | 1325 | — |
| LMArena Russian | 1385 | — |
| LMArena Spanish | 1386 | — |
Instruction Following Too close to call
Kimi K2 (Jul 2025): 71.1 (#156), Sonar: 71.4 (#150)
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |
| IFEval | 85% | — |
| LMArena Instruction Following | 1348 | — |
Long Context Not comparable
Kimi K2 (Jul 2025): 41.2 (#145), Sonar: —
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| Fiction.LiveBench | 66.7% | — |
| CL-bench | 17.6% | — |
| LMArena Longer Query | 1353 | — |
Writing & Preference Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 62.3 (#78), Sonar: 52.6 (#167)
| Benchmark | Kimi K2 (Jul 2025) | Sonar |
|---|---|---|
| LMArena Text | 1380 | — |
| LMArena Creative Writing | 1350 | — |
| Short-Story Creative Writing | 85.6% | — |
| EQ-Bench Creative Writing | 1666 | — |
| WildBench | 86.2% | — |
| LMArena Multi-Turn | 1371 | — |
| LiveBench Language | — | 44.1% |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Sonar?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.5 on the Noometry Index.
Which is cheaper, Kimi K2 (Jul 2025) or Sonar?
Sonar is cheaper. It lists at $1 per million input tokens and $1 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is Kimi K2 (Jul 2025) or Sonar better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 35.7 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 128K.
How many benchmarks do Kimi K2 (Jul 2025) and Sonar share?
0 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Sonar has 7.