DeepSeek · Model
DeepSeek V4.1 Flash 1017 Abliterated API
NewDeepSeek V4.1 Flash 1017 Abliterated answers at api.apmix.ai under the id below — one key and one monthly plan, on the OpenAI and Anthropic wire formats your tools already speak. No separate DeepSeek account, no per-provider bill.
Model id
deepseek/deepseek-v4.1-flash-1017-abliteratedStarterProMaxAt a glance
- Provider
- DeepSeek
- Included in
- Starter, Pro, Max — from $4.99/month
- Weighted-token cost
- 1× — a 1,000-token exchange costs 1K weighted tokens
- What a month buys
- Starter: 60M tokens · Pro: 200M tokens · Max: 500M tokens
- Context window
- 320K tokens
- Max output
- 131.1K tokens
- Reasoning
- Yes — thinks before it answers
- Tool calling
- Supported
- On apmix since
- November 21, 2026
Call DeepSeek V4.1 Flash 1017 Abliterated
Set the model to deepseek-v4.1-flash-1017-abliterated in any OpenAI- or Anthropic-compatible client. Both base URLs accept the same apmix key.
OpenAI-compatible · Chat Completions
curl https://api.apmix.ai/v1/chat/completions \
-H "Authorization: Bearer apx_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-v4.1-flash-1017-abliterated", "messages": [{"role": "user", "content": "Hello"}]}'Anthropic-compatible · Messages
curl https://api.apmix.ai/v1/messages \
-H "x-api-key: apx_live_YOUR_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-v4.1-flash-1017-abliterated", "max_tokens": 1024, "messages": [{"role": "user", "content": "Hello"}]}'Questions about DeepSeek V4.1 Flash 1017 Abliterated
- Which plan includes DeepSeek V4.1 Flash 1017 Abliterated?
- Starter, Pro, Max. Each plan is a monthly allowance of weighted tokens shared across every model and every key; DeepSeek V4.1 Flash 1017 Abliterated counts at 1× per token.
- How do I use DeepSeek V4.1 Flash 1017 Abliterated in Claude Code, Codex CLI or Cursor?
- Point the tool at apmix — https://api.apmix.ai/v1 for OpenAI-compatible clients, https://api.apmix.ai for Anthropic-compatible ones — with your apmix key, and set the model to deepseek-v4.1-flash-1017-abliterated. The Docs page has copy-paste setup for each tool.