Model comparison
Kimi K3 vs Muse Spark 1.2
Kimi K3 is the stronger model overall, scoring 59.5 to 50.3 on the Noometry Index. Muse Spark 1.2 costs 3.0× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
MetaSummary
- They share 30 benchmarks with published results for both. Kimi K3 scores higher in 8 categories and Muse Spark 1.2 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 46.4.
- The biggest single-benchmark swing is ProofBench: 87% for Kimi K3 and 43% for Muse Spark 1.2.
- Muse Spark 1.2 is cheaper at $1.25 / $4.25 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 has downloadable open weights; the other is API-only.
Side by side
| Kimi K3 | Muse Spark 1.2 | |
|---|---|---|
| Provider | Moonshot AI | Meta |
| Noometry Index | 59.5 | 50.3 |
| Released | 2026-07-16 | 2026-08-05 |
| Weights | Open | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 1.05M | 131K |
| Input $ / M tokens | $3 | $1.25 |
| Output $ / M tokens | $15 | $4.25 |
| Results tracked | 53 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Kimi K3: 61.0 (#10), Muse Spark 1.2: 49.2 (#51)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| DeepSWE | 68.5% | 54.9% |
| LMArena WebDev | 1654 | 1533 |
| FrontierSWE | 25.9% | 12% |
| SciCode | 59.5% | 56.4% |
| WeirdML | 82.6% | 60.3% |
| LMArena Coding | 1508 | 1495 |
| FrontierCode | 44.2% | — |
| ALE-Bench | 1,524 | — |
Agentic & Tool Use Kimi K3 leads
Kimi K3: 41.8 (#20), Muse Spark 1.2: 29.4 (#87)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| APEX-Agents | 50.6% | 36.4% |
| GDP.pdf | 19% | 16% |
| τ²-bench Banking | 37.1% | — |
| PostTrainBench | 32% | — |
| GBAEval | 48.3% | — |
| Vending-Bench 2 | 5,165 | — |
Reasoning Kimi K3 leads
Kimi K3: 63.0 (#17), Muse Spark 1.2: 51.3 (#34)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| SimpleBench | 60.7% | 74.5% |
| NYT Connections (extended) | 93.6% | 79.2% |
| CritPt | 23.4% | 17.7% |
| LMArena Hard Prompts | 1496 | 1486 |
| DTBench | 91.2% | 94.7% |
| LMCA | 52.7% | 48.4% |
| Epoch Capabilities Index | 157.45 | 154.87 |
| ARC-AGI-2 | 60.4% | — |
| ARC-AGI-1 | 94.5% | — |
| Chess Puzzles | 39% | — |
| Mystery Game Puzzles | 26% | — |
| Surface Evolver Bench | 95% | — |
| ForecastBench | 61.1 | — |
Math Kimi K3 leads
Kimi K3: 74.2 (#16), Muse Spark 1.2: 46.4 (#70)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| ProofBench | 87% | 43% |
| LMArena Math | 1491 | 1471 |
| FrontierMath (Tiers 1-3) | 72.2% | — |
| FrontierMath Tier 4 | 39% | — |
| MathArena Final-Answer Competitions | 87.8% | — |
| OTIS Mock AIME 2024-2025 | 97.2% | — |
Knowledge Kimi K3 leads
Kimi K3: 63.2 (#21), Muse Spark 1.2: 54.1 (#53)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| SimpleQA Verified | 50.6% | 60.3% |
| LMArena Expert | 1521 | 1480 |
| GPQA Diamond | 93.1% | — |
Multimodal Muse Spark 1.2 leads
Kimi K3: 37.8 (#70), Muse Spark 1.2: 43.4 (#25)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| LMArena Vision | — | 1305 |
| Blueprint-Bench 2 | 29.5% | — |
| Furniture Assembly | 34.2% | — |
Multilingual Too close to call
Kimi K3: 56.3 (#21), Muse Spark 1.2: 57.1 (#11)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| LMArena Non-English | 1466 | 1478 |
| LMArena Chinese | 1529 | 1511 |
| LMArena French | 1491 | 1513 |
| LMArena Russian | 1482 | 1487 |
| LMArena Spanish | 1472 | 1498 |
| LMArena German | 1488 | — |
| LMArena Japanese | 1487 | — |
| LMArena Korean | 1458 | — |
Instruction Following Kimi K3 leads
Kimi K3: 77.7 (#14), Muse Spark 1.2: 76.7 (#36)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| LMArena Instruction Following | 1483 | 1461 |
Long Context Too close to call
Kimi K3: 45.8 (#29), Muse Spark 1.2: 45.2 (#48)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| LMArena Longer Query | 1494 | 1475 |
Writing & Preference Kimi K3 leads
Kimi K3: 76.6 (#4), Muse Spark 1.2: 72.3 (#14)
| Benchmark | Kimi K3 | Muse Spark 1.2 |
|---|---|---|
| LMArena Text | 1476 | 1482 |
| LMArena Creative Writing | 1454 | 1449 |
| EQ-Bench Creative Writing | 2082 | 1840 |
| LMArena Multi-Turn | 1488 | 1494 |
| EQ-Bench 4 | 1339 | — |
Frequently asked questions
Is Kimi K3 better than Muse Spark 1.2?
Kimi K3 is the stronger model overall, scoring 59.5 to 50.3 on the Noometry Index. Muse Spark 1.2 costs 3.0× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Kimi K3 or Muse Spark 1.2?
Muse Spark 1.2 is cheaper. It lists at $1.25 per million input tokens and $4.25 per million output tokens; Kimi K3 lists at $3 and $15.
Is Kimi K3 or Muse Spark 1.2 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 49.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 1.05M tokens.
How many benchmarks do Kimi K3 and Muse Spark 1.2 share?
30 benchmarks have published results for both models. Kimi K3 has 53 scored results on Noometry and Muse Spark 1.2 has 31.