Model comparison
GLM-4.5-Air vs Kimi K2 (Jul 2025)
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.9 on the Noometry Index. GLM-4.5-Air costs 2.4× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. GLM-4.5-Air scores higher in 2 categories and Kimi K2 (Jul 2025) in 6 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in coding, where Kimi K2 (Jul 2025) leads 42.4 to 33.3.
- The biggest single-benchmark swing is Omni-MATH: 39.1% for GLM-4.5-Air and 65.4% for Kimi K2 (Jul 2025).
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) accepts more context: 262K tokens versus 131K.
Side by side
| GLM-4.5-Air | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Moonshot AI |
| Noometry Index | 38.9 | 41.2 |
| Released | 2025-07-20 | 2025-07-12 |
| Weights | Open | Open |
| Context window | 131K | 262K |
| Max output | 98K | 262K |
| Input $ / M tokens | $0.20 | $0.57 |
| Output $ / M tokens | $1.10 | $2.30 |
| Results tracked | 27 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2 (Jul 2025) leads
GLM-4.5-Air: 33.3 (#259), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| GSO | 2.9% | 4.9% |
| LMArena Coding | 1397 | 1399 |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| WeirdML | — | 42.8% |
| ALE-Bench | — | 597.5 |
Agentic & Tool Use Not comparable
GLM-4.5-Air: —, Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning Too close to call
GLM-4.5-Air: 24.1 (#166), Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| Kagi LLM Benchmark | 43% | 64.4% |
| LMArena Hard Prompts | 1379 | 1384 |
| ForecastBench | 59.2 | 60.2 |
| SimpleBench | — | 26.3% |
| Epoch Capabilities Index | — | 146.01 |
Math Kimi K2 (Jul 2025) leads
GLM-4.5-Air: 36.2 (#170), Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | 39.1% | 65.4% |
| LMArena Math | 1396 | 1397 |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Kimi K2 (Jul 2025) leads
GLM-4.5-Air: 35.0 (#191), Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | 76.2% | 81.9% |
| Vectara Hallucination Rate | 9.3% | 17.9% |
| GPQA (HELM) | 59.4% | 65.3% |
| LMArena Expert | 1370 | 1365 |
| Humanity's Last Exam | 8.1% | — |
| Confabulations | — | 20.4% |
Multilingual Too close to call
GLM-4.5-Air: 49.1 (#135), Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1366 | 1372 |
| LMArena Chinese | 1426 | 1415 |
| LMArena French | 1399 | 1379 |
| LMArena German | 1377 | 1387 |
| LMArena Japanese | 1348 | 1349 |
| LMArena Korean | 1308 | 1325 |
| LMArena Russian | 1373 | 1385 |
| LMArena Spanish | 1386 | 1386 |
Instruction Following Kimi K2 (Jul 2025) leads
GLM-4.5-Air: 69.6 (#171), Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | 81.2% | 85% |
| LMArena Instruction Following | 1354 | 1348 |
Long Context Too close to call
GLM-4.5-Air: 41.6 (#135), Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1366 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference Kimi K2 (Jul 2025) leads
GLM-4.5-Air: 55.9 (#139), Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | GLM-4.5-Air | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1384 | 1380 |
| LMArena Creative Writing | 1343 | 1350 |
| WildBench | 78.9% | 86.2% |
| LMArena Multi-Turn | 1371 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
Frequently asked questions
Is GLM-4.5-Air better than Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is the stronger model overall, scoring 41.2 to 38.9 on the Noometry Index. GLM-4.5-Air costs 2.4× less per token, which makes it the better buy when Kimi K2 (Jul 2025)'s lead doesn't matter for your workload.
Which is cheaper, GLM-4.5-Air or Kimi K2 (Jul 2025)?
GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is GLM-4.5-Air or Kimi K2 (Jul 2025) better for coding?
Kimi K2 (Jul 2025) scores higher on coding benchmarks: 42.4 versus 33.3 in the Noometry coding category.
Which has the bigger context window?
Kimi K2 (Jul 2025) does, with 262K tokens against 131K.
How many benchmarks do GLM-4.5-Air and Kimi K2 (Jul 2025) share?
26 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Kimi K2 (Jul 2025) has 42.