#3 of 6 on the Abliterated API Index
Morph review: abliterated API pricing and privacy
Morph scores 57.1 and ranks #3 of 6. One abliterated DeepSeek V4.1 Flash behind an approval step, with prompts kept up to 30 days.
CostCapabilityPrivacyAbliterated API Index
| Measure | Value | Rank | |
|---|---|---|---|
| Abliterated API Index | 57.1 | 3/6 | |
| Price | $0.525 / 1M | 2/6 | |
| Prompt retention | Up to 30 days | 6/6 | |
| Context window | 1M tokens | 1/6 | |
| 1M-context models | 1 model | 3/6 | |
| API formats | 10 / 10 | 1/6 | |
| Payment and access | 10 / 10 | 1/6 |
Read from Morph's own pages on November 18, 2026.Full review
At a glance
- Website
- morphllm.com
- Best model
- DeepSeek V4.1 Flash Uncensored
- Best-value 128K+ model
- DeepSeek V4.1 Flash Uncensored, $0.30 / $1.20
- API formats
- OpenAI Chat Completions, Anthropic Messages, OpenAI Responses
- Coding agents
- A Claude Code setup guide.
- Images
- The uncensored model takes images.
- Payment
- Prepaid credit from $10 by card or USDC, or a $200 Ultra plan.
- Catalog
- 1 model
Strengths
- A 1M-token context.
- OpenAI, Anthropic Messages and Responses endpoints.
- Takes USDC as well as card.
Where it falls short
- Morph keeps prompts and outputs for up to 30 days unless you sign an enterprise addendum.
- On this model, Morph keeps flagged prompts on every tier, zero-retention accounts included.
- Access needs approval before the first request.
- DeepSeek V4.1 Flash costs $0.30 in and $1.20 out per 1M tokens, against $0.24 and $0.90 at Silk Compute.
- One uncensored model, with its Kimi build listed as coming soon.
What it says about your prompts
"We keep Inputs and Outputs only to operate the Services and your own request history, delete them within 30 days."morphllm.com/privacy, read November 18, 2026
Prices
| Model | Input | Cached | Output | Context |
|---|---|---|---|---|
| DeepSeek V4.1 Flash Uncensored | $0.30 | $0.03 | $1.20 | 1M |
Per 1M tokens, from morphllm.com/pricing on November 18, 2026.
Visit Morph