New featureModels ·
GLM 5.3 and GLM 5.3 Flash
GLM 5.3GLM 5.3 Flash
What it does
Z.ai's latest pair joined the model selector — 5.3 for hard reasoning and tool use, 5.3 Flash for the same work at speed, with images on top.
Why it matters
A million tokens of context and Actions support on every plan, which is not a combination the frontier tiers offer.
The specs
Both are open models reached through OpenRouter, and they share a context budget — the split between them is speed and whether the model can see.
- Context
- 1,048,576 tokens in, up to 131,072 out, on both.
- Input
- GLM 5.3 is text only. GLM 5.3 Flash also takes images.
- Actions
- Supported on both.
- Reasoning
- Fast ($$), Smartest ($$$$), or Maximum ($$$$$) — they skip the middle tier.
- Plans
- Every plan.