Head-to-head showdown: GPT-5.3 Codex ($1.75 in / $14.00 out per 1M) vs GLM-4.6V (Z.ai) ($0.30 in / $0.90 out per 1M). GLM-4.6V (Z.ai) is 13.1× cheaper across standard token mixes, with 256K vs 128K context windows.
gpt-5.3-codex · openai
glm-4.6v · zai
Benchmark
| Workload Scenario | GPT-5.3 Codex | GLM-4.6V (Z.ai) | Price Delta |
|---|---|---|---|
| 1M input tokens (raw text) | $1.75 | $0.30 | +$1.45 |
| 1M output tokens (generation) | $14.00 | $0.90 | +$13.10 |
| 1M tokens · 70% input / 30% output mix | $5.425 | $0.48 | +$4.945 |
| Standard chat turn (4K in / 800 out, 50% cached) | $0.0151 | $0.00142 | +$0.0136 |
| Monthly scale (10K requests / day) | $4,515.00 | $426.00 | +$4,089.00 |
Negative difference = GPT-5.3 Codex is cheaper. Positive = GLM-4.6V (Z.ai) is cheaper.
Scaling Curve
Total cost of a token volume at 70/30 input/output split (uncached). Log-log scale.
Verdict
GLM-4.6V (Z.ai) 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 (GPT-5.3 Codex: 256K, GLM-4.6V (Z.ai): 128K) may justify the premium for your task.
FAQ
On input, GLM-4.6V (Z.ai) is cheaper ($0.30/M vs $1.75/M). On output, GLM-4.6V (Z.ai) is cheaper ($0.90/M vs $14.00/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 $0.0151 on GPT-5.3 Codex and $0.00142 on GLM-4.6V (Z.ai) — GLM-4.6V (Z.ai) is 10.6× more expensive for that workload.
GPT-5.3 Codex supports 256,000 tokens (64K max output); GLM-4.6V (Z.ai) supports 128,000 (32.8K max output). GPT-5.3 Codex fits 2.0× more context, which matters for long documents and agents.