| DeepSeek V4 Flash 0731 | GPT-5.6 Terra | |
|---|---|---|
| Input, per 1M tokens | $0.021 | $2.00 |
| Output, per 1M tokens | $0.32 | $12.00 |
| Context window | 1.31M | 1.05M |
| Classification or routing (600 tokens in, 10 out), per 1,000 | $0.0158 | $1.32 |
| Chat reply (2,000 in, 400 out), per 1,000 | $0.17 | $8.80 |
| Retrieval-augmented answer (8,000 in, 500 out), per 1,000 | $0.328 | $22.00 |
| Confidence signal | yes | no |
| Tool calling | yes | yes |
| BANKING77, correct of 400 | 298 | not measured |
Which one should carry your traffic?
If DeepSeek V4 Flash 0731 answers your requests as well as GPT-5.6 Terra, moving that traffic saves 98% on chat-shaped requests. Averages cannot tell you whether it does for your requests. DeepSeek V4 Flash 0731 exposes a confidence signal, so Gimbix can let it answer first and hand unsure cases to GPT-5.6 Terra.
DeepSeek V4 Flash 0731 profile · GPT-5.6 Terra profile · Measure it on your traffic