DeepSeek V4.1 Flash 1017 for AI agents
DeepSeek V4.1 Flash 1017 is the value model for an unrestricted AI agent, at $0.24 per million input tokens. A hundred agent steps cost $0.57 against $7.80 on Kimi K3.1 Pro.
First tool call
curl https://api.silkcompute.com/chat/completions \
-H "Authorization: Bearer $SILK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4.1-flash-1017-uncensored",
"messages": [{"role": "user", "content": "Is the disk on my server almost full?"}],
"tools": [{"type": "function", "function": {
"name": "run_command",
"description": "Run a shell command on my own server",
"parameters": {"type": "object",
"properties": {"cmd": {"type": "string"}}, "required": ["cmd"]}}}]
}'Specs
- Maker
- DeepSeek
- Parameters
- 763B
- Context
- 1,048,576 tokens
- Max output
- 131,072 tokens
- Input
- Text, images
- Per 1M tokens
- $0.24 input, $0.038 cached, $0.90 output
- Reasoning effort
- Low, high or max
- Model id
deepseek-v4.1-flash-1017-uncensored
One reply caps at 131,072 tokens, so split very long outputs across steps.
Cost of a run
Each step is 20,000 input and 1,000 output tokens, at prices read November 18, 2026.
50 steps
- No cache
- $0.29
- 80% cached
- $0.12
200 steps
- No cache
- $1.14
- 80% cached
- $0.49
1,000 steps
- No cache
- $5.70
- 80% cached
- $2.47
Frameworks
It runs in all 11 frameworks on this site. smolagents, Microsoft Agent Framework use it in their setup.
All 11 setups