A lower-refusal finetune of DeepSeek V4.1 Flash with text input, optional reasoning, and tool calling. Supports a context window of up to 1M tokens.
Specification
most-agreed values
Context
1M
Max output
393K
Released
2026-10-01
Knowledge cutoff
—
Retires
—
Open weights
yes
Input
text, image
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.33
Output
$1.32
Cache read
$0.0083
Available from 3 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| nano-gpt | $0.33 | $1.32 | $0.0083 | — | 1M | 393K | — |
| nano-gpt · thinking | $0.33 | $1.32 | $0.0083 | — | 1M | 393K | — |
| venice | $0.33 | $1.32 | $0.0083 | — | 1M | 131K | — |
More from DeepSeek
most-hosted first