Model comparison
Kimi K2 (Jul 2025) vs Muse Spark 1.2
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 2.0× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
MetaSummary
- They share 18 benchmarks with published results for both. Kimi K2 (Jul 2025) scores higher in 1 category and Muse Spark 1.2 in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Muse Spark 1.2 leads 51.3 to 23.3.
- The biggest single-benchmark swing is SimpleBench: 26.3% for Kimi K2 (Jul 2025) and 74.5% for Muse Spark 1.2.
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $1.25 / $4.25 for Muse Spark 1.2.
- Muse Spark 1.2 accepts more context: 1.05M tokens versus 262K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| Kimi K2 (Jul 2025) | Muse Spark 1.2 | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 41.2 | 50.3 |
| Released | 2025-07-12 | 2026-08-05 |
| Weights | Open | Proprietary |
| Context window | 262K | 1.05M |
| Max output | 262K | 131K |
| Input $ / M tokens | $0.57 | $1.25 |
| Output $ / M tokens | $2.30 | $4.25 |
| Results tracked | 42 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 42.4 (#102), Muse Spark 1.2: 49.2 (#51)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| WeirdML | 42.8% | 60.3% |
| LMArena Coding | 1399 | 1495 |
| DeepSWE | — | 54.9% |
| SWE-bench Verified (bash only) | 63.4% | — |
| Aider Polyglot | 59.1% | — |
| LMArena WebDev | — | 1533 |
| FrontierSWE | — | 12% |
| SciCode | — | 56.4% |
| GSO | 4.9% | — |
| ALE-Bench | 597.5 | — |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Kimi K2 (Jul 2025): 32.4 (#64), Muse Spark 1.2: 29.4 (#87)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| Terminal-Bench | 35.7% | — |
| APEX-Agents | — | 36.4% |
| Berkeley Function Calling Leaderboard | 59.1% | — |
| GDP.pdf | — | 16% |
| METR Time Horizons | 59.2% | — |
Reasoning Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 23.3 (#179), Muse Spark 1.2: 51.3 (#34)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| SimpleBench | 26.3% | 74.5% |
| LMArena Hard Prompts | 1384 | 1486 |
| Epoch Capabilities Index | 146.01 | 154.87 |
| Kagi LLM Benchmark | 64.4% | — |
| NYT Connections (extended) | — | 79.2% |
| CritPt | — | 17.7% |
| DTBench | — | 94.7% |
| LMCA | — | 48.4% |
| ForecastBench | 60.2 | — |
Math Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 42.7 (#83), Muse Spark 1.2: 46.4 (#70)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Math | 1397 | 1471 |
| ProofBench | — | 43% |
| Omni-MATH | 65.4% | — |
| FrontierMath (Feb 2025 set) | 21.4% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 37.3 (#157), Muse Spark 1.2: 54.1 (#53)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Expert | 1365 | 1480 |
| SimpleQA Verified | — | 60.3% |
| MMLU-Pro | 81.9% | — |
| Confabulations | 20.4% | — |
| Vectara Hallucination Rate | 17.9% | — |
| GPQA (HELM) | 65.3% | — |
Multimodal Not comparable
Kimi K2 (Jul 2025): —, Muse Spark 1.2: 43.4 (#25)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Vision | — | 1305 |
Multilingual Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 49.6 (#130), Muse Spark 1.2: 57.1 (#11)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Non-English | 1372 | 1478 |
| LMArena Chinese | 1415 | 1511 |
| LMArena French | 1379 | 1513 |
| LMArena Russian | 1385 | 1487 |
| LMArena Spanish | 1386 | 1498 |
| LMArena German | 1387 | — |
| LMArena Japanese | 1349 | — |
| LMArena Korean | 1325 | — |
Instruction Following Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 71.1 (#156), Muse Spark 1.2: 76.7 (#36)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Instruction Following | 1348 | 1461 |
| IFEval | 85% | — |
Long Context Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 41.2 (#145), Muse Spark 1.2: 45.2 (#48)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Longer Query | 1353 | 1475 |
| Fiction.LiveBench | 66.7% | — |
| CL-bench | 17.6% | — |
Writing & Preference Muse Spark 1.2 leads
Kimi K2 (Jul 2025): 62.3 (#78), Muse Spark 1.2: 72.3 (#14)
| Benchmark | Kimi K2 (Jul 2025) | Muse Spark 1.2 |
|---|---|---|
| LMArena Text | 1380 | 1482 |
| LMArena Creative Writing | 1350 | 1449 |
| EQ-Bench Creative Writing | 1666 | 1840 |
| LMArena Multi-Turn | 1371 | 1494 |
| Short-Story Creative Writing | 85.6% | — |
| WildBench | 86.2% | — |
Frequently asked questions
Is Kimi K2 (Jul 2025) better than Muse Spark 1.2?
Muse Spark 1.2 is the stronger model overall, scoring 50.3 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 2.0× less per token, which makes it the better buy when Muse Spark 1.2's lead doesn't matter for your workload.
Which is cheaper, Kimi K2 (Jul 2025) or Muse Spark 1.2?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Muse Spark 1.2 lists at $1.25 and $4.25.
Is Kimi K2 (Jul 2025) or Muse Spark 1.2 better for coding?
Muse Spark 1.2 scores higher on coding benchmarks: 49.2 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Muse Spark 1.2 does, with 1.05M tokens against 262K.
How many benchmarks do Kimi K2 (Jul 2025) and Muse Spark 1.2 share?
18 benchmarks have published results for both models. Kimi K2 (Jul 2025) has 42 scored results on Noometry and Muse Spark 1.2 has 31.