| gpt-oss-120b | GLM 5.3 | |
|---|---|---|
| Input, per 1M tokens | $0.15 | $0.325 |
| Output, per 1M tokens | $0.60 | $4.40 |
| Context window | 131K | 1.31M |
| Classification or routing (600 tokens in, 10 out), per 1,000 | $0.096 | $0.239 |
| Chat reply (2,000 in, 400 out), per 1,000 | $0.54 | $2.41 |
| Retrieval-augmented answer (8,000 in, 500 out), per 1,000 | $1.50 | $4.80 |
| Confidence signal | yes | yes |
| Tool calling | yes | yes |
| BANKING77, correct of 400 | 329 | not measured |
Which one should carry your traffic?
If gpt-oss-120b answers your requests as well as GLM 5.3, moving that traffic saves 78% on chat-shaped requests. Averages cannot tell you whether it does for your requests. gpt-oss-120b exposes a confidence signal, so Gimbix can let it answer first and hand unsure cases to GLM 5.3.
gpt-oss-120b profile · GLM 5.3 profile · Measure it on your traffic