Model comparison
Inkling vs Kimi K3
Kimi K3 is the stronger model overall, scoring 59.5 to 44.1 on the Noometry Index. Inkling costs 2.3× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. Inkling scores higher in 0 categories and Kimi K3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Kimi K3 leads 74.2 to 31.3.
- The biggest single-benchmark swing is ProofBench: 0% for Inkling and 87% for Kimi K3.
- Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $3 / $15 for Kimi K3.
- Kimi K3 accepts more context: 1.05M tokens versus 66K.
Side by side
| Inkling | Kimi K3 | |
|---|---|---|
| Provider | Thinking Machines Lab | Moonshot AI |
| Noometry Index | 44.1 | 59.5 |
| Released | 2026-07-15 | 2026-07-16 |
| Weights | Open | Open |
| Context window | 66K | 1.05M |
| Max output | 66K | 1.05M |
| Input $ / M tokens | $1.87 | $3 |
| Output $ / M tokens | $4.68 | $15 |
| Results tracked | 41 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K3 leads
Inkling: 34.5 (#234), Kimi K3: 61.0 (#10)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| FrontierCode | 14% | 44.2% |
| LMArena WebDev | 1413 | 1654 |
| FrontierSWE | 4.1% | 25.9% |
| SciCode | 47% | 59.5% |
| WeirdML | 32.3% | 82.6% |
| LMArena Coding | 1464 | 1508 |
| ALE-Bench | 946 | 1,524 |
| DeepSWE | — | 68.5% |
Agentic & Tool Use Kimi K3 leads
Inkling: 29.6 (#85), Kimi K3: 41.8 (#20)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| APEX-Agents | 33.8% | 50.6% |
| τ²-bench Banking | 25% | 37.1% |
| PostTrainBench | — | 32% |
| GBAEval | — | 48.3% |
| GDP.pdf | — | 19% |
| Vending-Bench 2 | — | 5,165 |
Reasoning Kimi K3 leads
Inkling: 40.4 (#56), Kimi K3: 63.0 (#17)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| ARC-AGI-2 | 36.5% | 60.4% |
| SimpleBench | 50% | 60.7% |
| ARC-AGI-1 | 79.5% | 94.5% |
| CritPt | 5.4% | 23.4% |
| Chess Puzzles | 21% | 39% |
| LMArena Hard Prompts | 1451 | 1496 |
| DTBench | 87.5% | 91.2% |
| LMCA | 37.6% | 52.7% |
| Epoch Capabilities Index | 148.54 | 157.45 |
| NYT Connections (extended) | — | 93.6% |
| Mystery Game Puzzles | — | 26% |
| Surface Evolver Bench | — | 95% |
| ForecastBench | — | 61.1 |
Math Kimi K3 leads
Inkling: 31.3 (#225), Kimi K3: 74.2 (#16)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 33.3% | 72.2% |
| FrontierMath Tier 4 | 4.9% | 39% |
| OTIS Mock AIME 2024-2025 | 88.9% | 97.2% |
| ProofBench | 0% | 87% |
| LMArena Math | 1479 | 1491 |
| MathArena Final-Answer Competitions | — | 87.8% |
Knowledge Kimi K3 leads
Inkling: 55.1 (#49), Kimi K3: 63.2 (#21)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| GPQA Diamond | 88.3% | 93.1% |
| SimpleQA Verified | 40.3% | 50.6% |
| LMArena Expert | 1465 | 1521 |
Multimodal Not comparable
Inkling: —, Kimi K3: 37.8 (#70)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| Blueprint-Bench 2 | — | 29.5% |
| Furniture Assembly | — | 34.2% |
Multilingual Kimi K3 leads
Inkling: 54.0 (#52), Kimi K3: 56.3 (#21)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| LMArena Non-English | 1434 | 1466 |
| LMArena Chinese | 1490 | 1529 |
| LMArena French | 1458 | 1491 |
| LMArena German | 1446 | 1488 |
| LMArena Japanese | 1429 | 1487 |
| LMArena Korean | 1404 | 1458 |
| LMArena Russian | 1429 | 1482 |
| LMArena Spanish | 1448 | 1472 |
Instruction Following Kimi K3 leads
Inkling: 75.1 (#71), Kimi K3: 77.7 (#14)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| LMArena Instruction Following | 1426 | 1483 |
Long Context Kimi K3 leads
Inkling: 43.8 (#86), Kimi K3: 45.8 (#29)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| LMArena Longer Query | 1434 | 1494 |
Writing & Preference Kimi K3 leads
Inkling: 65.2 (#51), Kimi K3: 76.6 (#4)
| Benchmark | Inkling | Kimi K3 |
|---|---|---|
| LMArena Text | 1441 | 1476 |
| LMArena Creative Writing | 1387 | 1454 |
| EQ-Bench Creative Writing | 1611 | 2082 |
| EQ-Bench 4 | 1226 | 1339 |
| LMArena Multi-Turn | 1436 | 1488 |
Frequently asked questions
Is Inkling better than Kimi K3?
Kimi K3 is the stronger model overall, scoring 59.5 to 44.1 on the Noometry Index. Inkling costs 2.3× less per token, which makes it the better buy when Kimi K3's lead doesn't matter for your workload.
Which is cheaper, Inkling or Kimi K3?
Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; Kimi K3 lists at $3 and $15.
Is Inkling or Kimi K3 better for coding?
Kimi K3 scores higher on coding benchmarks: 61.0 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K3 does, with 1.05M tokens against 66K.
How many benchmarks do Inkling and Kimi K3 share?
41 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Kimi K3 has 53.