DeepSeek
DeepSeek V3
Released 2024-12-26 · Knowledge up to 2024-07-31
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...
You send / 1M
$0.257
tokens in
It replies / 1M
$1.03
tokens out
Memory
164K
tokens it can read at once
Typical request
$0.0008
1,000 words in, 400 out
Data privacy
No training or retention: Yes. At least one provider serves it with zero data retention.
| Provider | You send / 1M | It replies / 1M | Uptime | Privacy policy |
|---|---|---|---|---|
| DeepInfra · fp4 | $0.256 | $0.712 | 100% | Read |
| DeepInfra · fp4 | $0.320 | $0.890 | 97% | Read |
Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.
What it can do
- Reads: text
- Writes: text
- Reasoning: No
- Tool calling: Yes
- Longest reply: 16K tokens
Similar models
- Qwen3 Coder 480B A35BQwen · $0.300 in / $1.00 out per 1M
- Qwen2.5 Coder 32B InstructQwen · $0.660 in / $1.00 out per 1M
- Hermes 3 405B InstructNousresearch · $1.00 in / $1.00 out per 1M
- Qwen3 Coder FlashQwen · $0.195 in / $0.975 out per 1M
- Qwen3 Next 80B A3B InstructQwen · $0.090 in / $1.10 out per 1M
- DeepSeek V3 0324DeepSeek · $0.290 in / $1.14 out per 1M