GPT-5.4 Mini vs GLM-5.3 Marathon
GLM-5.3 Marathon is cheaper on output tokens, while GLM-5.3 Marathon offers a larger context window. Choose GPT-5.4 Mini or GLM-5.3 Marathon based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | GPT-5.4 Mini | GLM-5.3 Marathon |
|---|---|---|
| Provider | OpenAI | Subconscious |
| Input / 1M tokens | $0.75 | $1.40 |
| Output / 1M tokens | $4.50 | $4.40 |
| Context window | 400K | 5M |
| Parameters | — | — |
| Open weights | No | Yes |
| Released | Mar 2026 | Aug 2026 |
| LiveBench score | 67.7 | — |
| GPQA Diamond score | 86.9 | — |
| FrontierMath score | 28.3 | — |
| Epoch Capabilities Index score | 149 | — |
| LMArena (Chatbot Arena) Elo score | 1412 | — |
| LMArena WebDev Elo score | 1398 | — |
| LMArena Vision Elo score | 1245 | — |
FAQ
Is GPT-5.4 Mini or GLM-5.3 Marathon cheaper?
GLM-5.3 Marathon is cheaper on output tokens ($4.40 vs $4.50 per 1M).
Which has the larger context window, GPT-5.4 Mini or GLM-5.3 Marathon?
GLM-5.3 Marathon has the larger context window (5M tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.