AI models
MiniMax M3 vs GLM-5.3
Same criteria, side by side. Where the gap is small, it may be measurement noise.
Also searched as
MiniMax M vs GLM-5
Direct answer: GPQA Diamond: MiniMax M3 90.9%, GLM-5.3 90.9% (Epoch AI). OTIS Mock AIME: MiniMax M3 71.1%, GLM-5.3 91.1% (Epoch AI). GPQA Diamond (OpenRouter): MiniMax M3 91.3%, GLM-5.3 87.9% (OpenRouter Benchmarks). τ-bench Verified (airline): MiniMax M3 76.3%, GLM-5.3 78.8% (OpenRouter Benchmarks).
| Metric | MiniMax M3 | GLM-5.3 | Source |
|---|---|---|---|
| Weights ?what it is | open | open | no data |
| Input US$/1M ? | US$ 0.30 | US$ 0.07 | no data |
| Output US$/1M ? | US$ 1.20 | US$ 1.40 | no data |
| Context ? | 1,048,576 | 1,048,576 | no data |
| GPQA Diamond ? | 90.9% | 90.9% | Epoch AI |
| OTIS Mock AIME ? | 71.1% | 91.1% | Epoch AI |
| FrontierMath (tiers 1 to 3) ? | no data | 68.8% | Epoch AI |
| SimpleQA Verified ? | no data | 41% | Epoch AI |
| GPQA Diamond (OpenRouter) ? | 91.3% | 87.9% | OpenRouter Benchmarks |
| τ-bench Verified (airline) ? | 76.3% | 78.8% | OpenRouter Benchmarks |
Compare with
See lab: MiniMax · Zhipu AI (Z.ai)
Sources
- Epoch AI, CC BY 4.0, latest version we checked.
- OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
- OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.