Model comparison
GLM-5.3-Flash vs Kimi K2 (Jul 2025)
GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 40.9 on the Noometry Index.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. GLM-5.3-Flash scores higher in 9 categories and Kimi K2 (Jul 2025) in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GLM-5.3-Flash leads 58.4 to 36.9.
- GLM-5.3-Flash is cheaper at $0.15 / $0.50 per million input/output tokens, against $0.57 / $2.30 for Kimi K2 (Jul 2025).
- GLM-5.3-Flash accepts more context: 1M tokens versus 262K.
Side by side
| GLM-5.3-Flash | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Moonshot AI |
| Noometry Index | 49.6 | 40.9 |
| Released | 2026-08-20 | 2025-07-12 |
| Weights | Open | Open |
| Context window | 1M | 262K |
| Max output | 131K | 262K |
| Input $ / M tokens | $0.15 | $0.57 |
| Output $ / M tokens | $0.50 | $2.30 |
| Results tracked | 36 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3-Flash leads
GLM-5.3-Flash: 53.7 (#26), Kimi K2 (Jul 2025): 43.1 (#83)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Coding | 1506 | 1399 |
| ALE-Bench | 303.55 | 597.5 |
| DeepSWE | 63.4% | — |
| FrontierCode | 31.8% | — |
| SWE-bench Verified (bash only) | — | 63.4% |
| Aider Polyglot | — | 59.1% |
| LMArena WebDev | 1614 | — |
| SciCode | 51.6% | — |
| GSO | — | 4.9% |
| WeirdML | — | 42.8% |
Agentic & Tool Use GLM-5.3-Flash leads
GLM-5.3-Flash: 34.3 (#41), Kimi K2 (Jul 2025): 32.2 (#59)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | — | 35.7% |
| APEX-Agents | 52.8% | — |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| GDP.pdf | 14% | — |
| METR Time Horizons | — | 59.2% |
Reasoning GLM-5.3-Flash leads
GLM-5.3-Flash: 31.7 (#69), Kimi K2 (Jul 2025): 22.7 (#173)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Hard Prompts | 1490 | 1384 |
| Epoch Capabilities Index | 151.9 | 146.02 |
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| CritPt | 15.4% | — |
| Chess Puzzles | 14% | — |
| Mystery Game Puzzles | 8% | — |
| Surface Evolver Bench | 52.5% | — |
| Bench to the Future 3 | 0.15 | — |
| ForecastBench | — | 60.2 |
Math GLM-5.3-Flash leads
GLM-5.3-Flash: 53.1 (#38), Kimi K2 (Jul 2025): 40.1 (#81)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Math | 1499 | 1396 |
| FrontierMath (Tiers 1-3) | 55.8% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 93.9% | — |
| ProofBench | 21% | — |
| Omni-MATH | — | 65.4% |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge GLM-5.3-Flash leads
GLM-5.3-Flash: 58.4 (#33), Kimi K2 (Jul 2025): 36.9 (#153)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Expert | 1512 | 1363 |
| GPQA Diamond | 90.2% | — |
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
Multimodal Not comparable
GLM-5.3-Flash: 42.6 (#25), Kimi K2 (Jul 2025): —
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Vision | 1304 | — |
Multilingual GLM-5.3-Flash leads
GLM-5.3-Flash: 56.2 (#24), Kimi K2 (Jul 2025): 49.8 (#125)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | 1462 | 1372 |
| LMArena Chinese | 1529 | 1416 |
| LMArena French | 1495 | 1379 |
| LMArena German | 1457 | 1386 |
| LMArena Japanese | 1420 | 1349 |
| LMArena Korean | 1443 | 1324 |
| LMArena Russian | 1472 | 1384 |
| LMArena Spanish | 1471 | 1383 |
Instruction Following GLM-5.3-Flash leads
GLM-5.3-Flash: 77.8 (#17), Kimi K2 (Jul 2025): 71.3 (#151)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Instruction Following | 1480 | 1347 |
| IFEval | — | 85% |
Long Context GLM-5.3-Flash leads
GLM-5.3-Flash: 45.3 (#41), Kimi K2 (Jul 2025): 41.2 (#140)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Longer Query | 1479 | 1353 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
Writing & Preference GLM-5.3-Flash leads
GLM-5.3-Flash: 65.4 (#49), Kimi K2 (Jul 2025): 62.5 (#76)
| Benchmark | GLM-5.3-Flash | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | 1471 | 1380 |
| LMArena Creative Writing | 1440 | 1349 |
| LMArena Multi-Turn | 1471 | 1371 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |
Frequently asked questions
Is GLM-5.3-Flash better than Kimi K2 (Jul 2025)?
GLM-5.3-Flash is the stronger model overall, scoring 49.6 to 40.9 on the Noometry Index.
Which is cheaper, GLM-5.3-Flash or Kimi K2 (Jul 2025)?
GLM-5.3-Flash is cheaper. It lists at $0.15 per million input tokens and $0.50 per million output tokens; Kimi K2 (Jul 2025) lists at $0.57 and $2.30.
Is GLM-5.3-Flash or Kimi K2 (Jul 2025) better for coding?
GLM-5.3-Flash scores higher on coding benchmarks: 53.7 versus 43.1 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3-Flash does, with 1M tokens against 262K.
How many benchmarks do GLM-5.3-Flash and Kimi K2 (Jul 2025) share?
19 benchmarks have published results for both models. GLM-5.3-Flash has 36 scored results on Noometry and Kimi K2 (Jul 2025) has 42.