DeepSeek V3.1
DeepSeek
DeepSeek-V3.1 is post-trained on the top of DeepSeek-V3.1-Base, which is built upon the original V3 base checkpoint through a two-phase long context extension approach, following the methodology outlined in the original DeepSeek-V3 report. We have expanded our dataset by collecting additional long documents and substantially extending both training phases. The 32K extension phase has been increased 10-fold to 630B tokens, while the 128K extension phase has been extended by 3.3x to 209B tokens. Additionally, DeepSeek-V3.1 is trained using the UE8M0 FP8 scale data format to ensure compatibility with microscaling data formats.
- Language
- Implicit caching
- Reasoning
- Tool use
- Context
- 164K
- Max output
- 128K
- Input / 1M
- $0.250
- Output / 1M
- $0.950
Quickstart
Point any OpenAI-compatible client at Maxynetic Systems and pass deepseek/deepseek-v3.1 as the model.
curl https://api.maxynetic.com/v1/chat/completions \
-H "Authorization: Bearer $MAXYNETIC_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v3.1",
"messages": [{ "role": "user", "content": "Explain routing in one line." }]
}'Pricing
Billed per token with no markup on the underlying provider rate.
Input
Per 1M prompt tokens
$0.250
Output
Per 1M completion tokens
$0.950
Cache read
Per 1M cached prompt tokens
$0.130
A 1M-token prompt with 250K tokens of output costs about $0.487.
Rates for this model vary by upstream provider; Maxynetic Systems routes to the cheapest healthy one by default.