DeepSeek
DeepSeek V4.1 Flash 1017 Abliterated
DeepSeek V4.1 Flash 1017 Abliterated is DeepSeek's 763B October checkpoint with its refusal direction removed, and the best-value model in this ranking.
Rank#2 / 8
#2Editors' rankingSize#2 / 8
763BTotal parametersPrice#1 / 5
$0.405Blended per 1M, lowest hostContext
1MTokens per requestOur pick for coding agents and all-day use.
Input
Price in / out$0.24 / $0.90
Context1M tokens
Max output131,072
ReleasedOctober 2026
Model id
deepseek-v4.1-flash-1017-uncensoredIts low price counts most in agent runs, which resend the whole context every turn.
Silk Compute serves it with low, high and max reasoning effort, so triage can run on low and a hard bug on max.
Where to run DeepSeek V4.1 Flash by API
Blended price per 1M tokens, 3 input to 1 output.
| Provider | Model id | Input | Cached | Output | Context | Prompts |
|---|---|---|---|---|---|---|
| Silk Compute | deepseek-v4.1-flash-1017-uncensored | $0.24 | $0.038 | $0.90 | 1M | Never stored |
| MorphApproval required | morph-dsv41flash-uncensored | $0.30 | $0.03 | $1.20 | 1M | Up to 30 days |
| abliteration.ai | abliterated-model-large-flash | $1.00 | $0.10 | $2.00 | 1M | Never stored |
Per 1M tokens, read from pricing pages on November 18, 2026.
Try DeepSeek V4.1 Flash in two lines
Set the base URL and your key in the OpenAI or Anthropic SDK.
from openai import OpenAI client = OpenAI( base_url="https://api.silkcompute.com", api_key="YOUR_API_KEY", ) reply = client.chat.completions.create( model="deepseek-v4.1-flash-1017-uncensored", messages=[{"role": "user", "content": "..."}], )
In Claude Code:
export ANTHROPIC_BASE_URL=https://api.silkcompute.com export ANTHROPIC_AUTH_TOKEN=YOUR_API_KEY export ANTHROPIC_MODEL=deepseek-v4.1-flash-1017-uncensored claudeDeepSeek V4.1 Flash on Silk Compute