| Hy4 preview | GLM 5.3 Flash | |
|---|---|---|
| Input, per 1M tokens | $0.834 | $0.15 |
| Output, per 1M tokens | $2.50 | $0.50 |
| Context window | 1.05M | 1.31M |
| Classification or routing (600 tokens in, 10 out), per 1,000 | $0.525 | $0.095 |
| Chat reply (2,000 in, 400 out), per 1,000 | $2.67 | $0.50 |
| Retrieval-augmented answer (8,000 in, 500 out), per 1,000 | $7.92 | $1.45 |
| Confidence signal | no | yes |
| Tool calling | yes | yes |
| BANKING77, correct of 400 | not measured | 320 |
Which one should carry your traffic?
If GLM 5.3 Flash answers your requests as well as Hy4 preview, moving that traffic saves 81% on chat-shaped requests. Averages cannot tell you whether it does for your requests. GLM 5.3 Flash exposes a confidence signal, so Gimbix can let it answer first and hand unsure cases to Hy4 preview.
Hy4 preview profile · GLM 5.3 Flash profile · Measure it on your traffic