Model comparison
Inkling vs Kimi K2 (Jul 2025)
Inkling is the stronger model overall, scoring 44.1 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 2.6× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Summary
- They share 22 benchmarks with published results for both. Inkling scores higher in 6 categories and Kimi K2 (Jul 2025) in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.1 to 37.3.
- The biggest single-benchmark swing is SimpleBench: 50% for Inkling and 26.3% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 66K.
Side by side
| Inkling | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Thinking Machines Lab | Moonshot AI |
| Noometry Index | 44.1 | 41.2 |
| Released | 2026-07-15 | 2025-07-12 |
| Weights | Open | Open |
| Context window | 66K | 262K |
| Max output | 66K | 262K |
| Input $ / M tokens | $1.87 | $0.57 |
| Output $ / M tokens | $4.68 | $2.30 |
| Results tracked | 41 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
Inkling: 34.5 (#234), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| WeirdML | 32.3% | 42.8% |
| LMArena Coding | 1464 | 1399 |
| ALE-Bench | 946 | 597.5 |
| FrontierCode | 14% | — |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| LMArena WebDev | 1413 | — |
| FrontierSWE | 4.1% | — |
| SciCode | 47% | — |
| GSO | — | 4.9% |
Agentic & Tool Use Kimi K2 (Jul 2025) leads
Inkling: 29.6 (#85), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| APEX-Agents | 33.8% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| τ²-bench Banking | 25% | — |
| METR Time Horizons | — | 59.2% |
Reasoning Inkling leads
Inkling: 40.4 (#56), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | 50% | 26.3% |
| LMArena Hard Prompts | 1451 | 1384 |
| Epoch Capabilities Index | 148.54 | 146.01 |
| ARC-AGI-2 | 36.5% | — |
| Kagi LLM Benchmark | — | 64.4% |
| ARC-AGI-1 | 79.5% | — |
| CritPt | 5.4% | — |
| Chess Puzzles | 21% | — |
| DTBench | 87.5% | — |
| LMCA | 37.6% | — |
| ForecastBench | — | 60.2 |
Math Kimi K2 (Jul 2025) leads
Inkling: 31.3 (#225), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1479 | 1397 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Inkling leads
Inkling: 55.1 (#49), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1465 | 1365 |
| GPQA Diamond | 88.3% | — |
| SimpleQA Verified | 40.3% | — |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1434 | 1372 |
| LMArena Chinese | 1490 | 1415 |
| LMArena French | 1458 | 1379 |
| LMArena German | 1446 | 1387 |
| LMArena Japanese | 1429 | 1349 |
| LMArena Korean | 1404 | 1325 |
| LMArena Russian | 1429 | 1385 |
| LMArena Spanish | 1448 | 1386 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1426 | 1348 |
| IFEval | — | 85% |
Long Context Inkling leads
Inkling: 43.8 (#86), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1434 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | Inkling | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1441 | 1380 |
| LMArena Creative Writing | 1387 | 1350 |
| EQ-Bench Creative Writing | 1611 | 1666 |
| LMArena Multi-Turn | 1436 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| WildBench | — | 86.2% |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Kimi K2 (Jul 2025)?
Inkling is the stronger model overall, scoring 44.1 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 2.6× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, Inkling or Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; Inkling lists at $1.87 and $4.68.
Is Inkling or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 34.5 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 66K.
How many benchmarks do Inkling and Kimi K2 (Jul 2025) share?
22 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Kimi K2 (Jul 2025) has 42.