| Command A | DeepSeek V4 Flash 0731 | |
|---|---|---|
| Input, per 1M tokens | $2.50 | $0.021 |
| Output, per 1M tokens | $10.00 | $0.32 |
| Context window | 256K | 1.31M |
| Classification or routing (600 tokens in, 10 out), per 1,000 | $1.60 | $0.0158 |
| Chat reply (2,000 in, 400 out), per 1,000 | $9.00 | $0.17 |
| Retrieval-augmented answer (8,000 in, 500 out), per 1,000 | $25.00 | $0.328 |
| Confidence signal | no | yes |
| Tool calling | no | yes |
| BANKING77, correct of 400 | not measured | 298 |
Which one should carry your traffic?
If DeepSeek V4 Flash 0731 answers your requests as well as Command A, 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 Command A.
Command A profile · DeepSeek V4 Flash 0731 profile · Measure it on your traffic