Head-to-head showdown: GLM-4.5 (Z.ai) ($0.60 in / $2.20 out per 1M) vs GLM 4.7 (Zhipu) ($0.60 in / $2.20 out per 1M). GLM-4.5 (Z.ai) is 1.0× cheaper across standard token mixes, with 128K vs 200K context windows.
glm-4.5 · zai
glm-4.7 · zai
Benchmark
| Workload Scenario | GLM-4.5 (Z.ai) | GLM 4.7 (Zhipu) | Price Delta |
|---|---|---|---|
| 1M input tokens (raw text) | $0.60 | $0.60 | — |
| 1M output tokens (generation) | $2.20 | $2.20 | — |
| 1M tokens · 70% input / 30% output mix | $1.08 | $1.08 | — |
| Standard chat turn (4K in / 800 out, 50% cached) | $0.00318 | $0.00318 | — |
| Monthly scale (10K requests / day) | $954.00 | $954.00 | — |
Negative difference = GLM-4.5 (Z.ai) is cheaper. Positive = GLM 4.7 (Zhipu) is cheaper.
Scaling Curve
Total cost of a token volume at 70/30 input/output split (uncached). Log-log scale.
Verdict
GLM 4.7 (Zhipu) is cheaper on both input and output rates, so it costs less at every input/output mix. Price alone still isn't the whole decision: capability, latency and context limits (GLM-4.5 (Z.ai): 128K, GLM 4.7 (Zhipu): 200K) may justify the premium for your task.
FAQ
On input, GLM 4.7 (Zhipu) is cheaper ($0.60/M vs $0.60/M). On output, GLM-4.5 (Z.ai) is cheaper ($2.20/M vs $2.20/M). The same model is cheaper on both sides, so it wins at every mix.
A chat-style request (4,000 input + 800 output tokens, 50% cached) costs $0.00318 on GLM-4.5 (Z.ai) and $0.00318 on GLM 4.7 (Zhipu) — GLM-4.5 (Z.ai) is equal for that workload.
GLM-4.5 (Z.ai) supports 128,000 tokens (98.3K max output); GLM 4.7 (Zhipu) supports 200,000 (131.1K max output). GLM 4.7 (Zhipu) fits 1.6× more context, which matters for long documents and agents.