GLM-5.3 vs Kimi K3
GLM-5.3 (Z.ai (Zhipu AI)) and Kimi K3 (Moonshot) compared on benchmarks, pricing, context window, and use-case rankings.
- Kimi K3 leads on overall intelligence (ECI 159 vs 157).
- Kimi K3 is cheaper on input tokens: $0.28 vs $0.70 per 1M.
- Kimi K3 ranks higher for coding (93rd vs 88th percentile).
| Attribute | GLM-5.3 | Kimi K3 |
|---|---|---|
| Scores | ||
| Intelligence (ECI) | 157 | 159 |
| Coding | 88 | 93 |
| Math | 34 | 58 |
| Reasoning & Knowledge | 61 | 63 |
| Agentic & Tools | 70 | 50 |
| Specifications | ||
| Developer | Z.ai (Zhipu AI) | Moonshot |
| Family | — | Kimi |
| Released | Aug 14, 2026 | Jul 16, 2026 |
| Parameters | 744B | 2.8T |
| Availability | Open weights (restricted use) | Open weights (non-commercial) |
| Context window | 1M | 1M |
| Price — $/M input | $0.70 | $0.28 |
| Price — $/M output | $1.90 | $9.94 |
| Inputs | text | text, image, video |
| Outputs | text | text |
| Benchmarks | ||
| AIME 2024/2025 | 91% | 97% |
| APEX | 57% | 48% |
| GPQA Diamond | 91% | 93% |
| Humanity's Last Exam | 42% | 47% |
| SciCode | 59% | 60% |
| SimpleBench | 66% | 61% |
| SimpleQA Verified | 41% | 51% |
| WebDev Arena | 1614 | 1674 |
| WeirdML | 75% | 83% |
| ARC-AGI | — | 95% |
| ARC-AGI-2 | — | 60% |
Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.
Frequently asked questions
Is GLM-5.3 better than Kimi K3?
On Epoch AI's Capabilities Index, Kimi K3 scores higher (159) than GLM-5.3 (157). The right pick depends on your task, so compare their coding, math, and reasoning scores in the table above.
Which is cheaper, GLM-5.3 or Kimi K3?
Kimi K3 is cheaper on input tokens at $0.28 per million, versus $0.70 (representative OpenRouter pricing).
Which has a larger context window, GLM-5.3 or Kimi K3?
Neither: both support up to 1M tokens.
Which is better for coding, GLM-5.3 or Kimi K3?
Kimi K3 ranks higher across coding benchmarks like WebDev Arena and SciCode: 93rd vs 88th percentile among the models tracked on Model Beat.
Want a different match-up? Open the compare tool to add or swap models.
More comparisons
Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.