GLM API pricing: every model, every rate
GLM has moved quickly through versions. Some legacy entries still carry much higher rates than current models — worth checking before you migrate.
Full rate card
32 billable models. 0 carry a rate verified against the vendor's own published pricing; the rest are gateway rates only. USD per 1M tokens, verified 2026-09-16.
| Model | Input | Output | Out/In | Official (in / out) |
|---|---|---|---|---|
glm-4-alltools | $120 | — | not verified | |
glm-4-0520 | $60 | — | not verified | |
glm-4-plus | $30 | — | not verified | |
chatglm_pro | $10 | — | not verified | |
glm-4v-plus | $6 | — | not verified | |
chatglm_std | $5 | — | not verified | |
chatglm_turbo | $5 | — | not verified | |
glm-5-turbo | $2.5 | $11 | 4.4× | not verified |
chatglm_lite | $2 | — | not verified | |
glm-4.5-airx | $2 | $8 | 4.0× | not verified |
glm-4v | $1.5 | $6 | 4.0× | not verified |
glm-4.5v | $1 | $3 | 3.0× | not verified |
glm-4.5-x | $0.8 | $3.2 | 4.0× | not verified |
glm-3-turbo | $0.75 | — | not verified | |
glm-5.1 | $0.7 | $2.2 | 3.1× | not verified |
glm-5.2 | $0.7 | $2.2 | 3.1× | not verified |
glm-5.3 | $0.7 | $2.2 | 3.1× | not verified |
glm-4.6v | $0.5 | $1.5 | 3.0× | not verified |
glm-5 | $0.5 | $2.25 | 4.5× | not verified |
glm-4.7 | $0.35 | $1.63333 | 4.7× | not verified |
glm-4.7-thinking | $0.35 | $1.4 | 4.0× | not verified |
glm-4.5 | $0.3 | $1.2 | 4.0× | not verified |
glm-4.6 | $0.3 | $1.2 | 4.0× | not verified |
glm-4.6-thinking | $0.3 | $1.2 | 4.0× | not verified |
glm-4-air | $0.15 | — | not verified | |
glm-4.5-air | $0.1 | $0.55 | 5.5× | not verified |
glm-5.3-flash | $0.075 | $0.25 | 3.3× | not verified |
glm-4 | $0.0265 | $0.106 | 4.0× | not verified |
glm-4-airx | $0.0265 | $0.106 | 4.0× | not verified |
glm-4-flash | $0.0265 | $0.106 | 4.0× | not verified |
glm-4-long | $0.0265 | $0.106 | 4.0× | not verified |
glm-4.5-flash | $0.01 | $0.04 | 4.0× | not verified |
Output is billed at a multiple of input on most models. The Out/In column is that multiple — it matters more than the input price once output dominates your bill.
Across 32 billable models the rate spans $0.01 (glm-4.5-flash) to $120 (glm-4-alltools) per million input tokens — a 12000× spread. Picking the wrong tier is usually the single most expensive mistake here.
What Zhipu is good at
GLM-5.x is a strong general-purpose Chinese-language model with competitive agentic performance.
See all 827 models → or use the cost calculator with your own token split.