Model registry
The unrestricted registry.
Every uncensored, abliterated and security-tuned model Ophiris serves, behind one OpenAI-compatible endpoint. Compare capabilities, context windows and reference pricing, then see which models get used most and how they score on public benchmarks.
58 models
Rank 1 this week: Abliterated Model Large V2abliterationCapabilities: Reasoning, Tools, Structured output1M$3.00$5.00
Abliteration.ai's default unrestricted large text reasoning model is derived from GLM-5.3 and weight-modified to reduce refusals compared with the base model. It supports native tool calling, structured output, automatic prompt caching, and a one-million-token context window.
- Model ID
- Tier
- Frontier
- Context
- 1,000,000 tokens
- Max output
- 999,990 tokens
- Added
- Aug 31
Rank 2 this week: GLM 5.3 UncensoredZ.aiCapabilities: Reasoning, Tools1M$1.25$2.25
An experimental uncensored version of the full GLM 5.3 reasoning model for chat, creative writing, coding, and tool use. It may show unusual behavior or inconsistent response quality.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 131,072 tokens
- Added
- Sep 26
Rank 3 this week: GLM 5.3 Flash UncensoredZ.aiCapabilities: Reasoning, Tools, Vision, Structured output1M$0.20$0.80
GLM 5.3 Flash Uncensored is an uncensored fine-tune of the efficient 320B mixture-of-experts reasoning model with provider-dependent vision support, built for unrestricted chat, creative writing, coding, agentic work, tool use, and long-context tasks.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 32,768 tokens
- Added
- Jul 29
Rank 4 this week: DeepSeek V4.1 Flash 1017 UncensoredDeepSeekCapabilities: Reasoning, Tools, Vision1M$0.33$1.32
Uncensored DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. The date label is retained for compatibility. Served only by its own checkpoint; it may be unavailable while that host is down.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Oct 27
Rank 5 this week: DeepSeek V4.1 Flash 1017 Uncensored ThinkingDeepSeekCapabilities: Reasoning, Tools, Vision1M$0.33$1.32
Uncensored DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. The date label is retained for compatibility. Served only by its own checkpoint; it may be unavailable while that host is down.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Oct 27
Rank 6 this week: Abliterated Large FlashabliterationCapabilities: Reasoning, Tools, Vision, Structured output, Parallel tools1M$1.00$2.00
Abliterated Large Flash is an uncensored reasoning model based on DeepSeek V4.1-Flash-1017. It accepts text and images and returns text, with tool calling, JSON mode, JSON Schema output, and automatic prompt caching. Its shared input/output context is 1,048,576 tokens, with up to 262,144 output tokens when space permits. Reasoning defaults to high and can be turned off or set to low, high, or max. Supports up to four images across the request history.
- Model ID
- Tier
- Efficient
- Context
- 1,048,576 tokens
- Max output
- 262,144 tokens
- Added
- Oct 23
Rank 7 this week: MiMo V2.6 Flash AbliteratedxiaomiCapabilities: Reasoning, Tools, Vision1M$0.50$1.50
A MiMo V2.6 Flash finetune with refusal-direction ablation, image input, a 1M-token context window, separate reasoning output, and tool calling.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Sep 25
Rank 8 this week: DeepSeek V4.1 Flash Heretic ThinkingpqhazCapabilities: Reasoning, Tools, Vision512K$0.198$0.792
pqhaz's Heretic checkpoint of DeepSeek V4.1 Flash 1017, served by Sable at original mixed precision (FP8 dense/Engram and MXFP4 experts). Supports text and image input, tools, structured output, and optional reasoning. Separate from the DeAlign uncensored checkpoint.
- Model ID
- Tier
- Open weights
- Context
- 512,000 tokens
- Max output
- 128,000 tokens
- Added
- Oct 31
Rank 9 this week: DeepSeek V4.1 Flash Uncensored ThinkingDeepSeekCapabilities: Reasoning, Tools, Vision1M$0.33$1.32
A lower-refusal finetune of DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. Supports a context window of up to 1M tokens.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Oct 1
Rank 10 this week: Kimi K3.1 Pro Abliterated (Lyber)lyberCapabilities: Reasoning, Tools, Structured output1M$4.00$15.00
Kimi K3.1 Pro Abliterated served by Lyber.ai.
- Model ID
- Tier
- Frontier
- Context
- 1,048,576 tokens
- Max output
- 1,048,576 tokens
- Added
- Yesterday
Rank 11 this week: DeepSeek V4.1 Flash 0910 Uncensored ThinkingDeepSeekCapabilities: Reasoning, Tools, Vision524.3K$0.33$1.32
Uncensored DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. The date label is retained for compatibility. Served only by its own checkpoint; it may be unavailable while that host is down.
- Model ID
- Tier
- Open weights
- Context
- 524,288 tokens
- Max output
- 32,768 tokens
- Added
- Oct 27
Rank 12 this week: DeepSeek V4.1 Flash UncensoredDeepSeekCapabilities: Reasoning, Tools, Vision1M$0.33$1.32
A lower-refusal finetune of DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. Supports a context window of up to 1M tokens.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Oct 1
Rank 13 this week: Qwen 3.8 27B UncensoredQwenCapabilities: Reasoning, Tools, Vision, Structured output1M$1.00$3.00
Qwen 3.8 27B Uncensored is an FP8 open-weight multimodal model LoRA-tuned for fewer refusals across chat, coding, reasoning, tool use, and long-context work.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Jul 29
Rank 14 this week: Abliterated Model LargeabliterationCapabilities: Reasoning, Tools, Structured output1M$3.00$5.00
Abliteration.ai's unrestricted large text reasoning model is derived from GLM-5.2 and weight-modified to reduce refusals compared with the base model. It supports native tool calling, structured output, automatic prompt caching, and a one-million-token context window.
- Model ID
- Tier
- Frontier
- Context
- 1,000,000 tokens
- Max output
- 999,990 tokens
- Added
- Aug 27
Rank 15 this week: DeepSeek V4.1 Flash HereticpqhazCapabilities: Reasoning, Tools, Vision512K$0.198$0.792
pqhaz's Heretic checkpoint of DeepSeek V4.1 Flash 1017, served by Sable at original mixed precision (FP8 dense/Engram and MXFP4 experts). Supports text and image input, tools, structured output, and optional reasoning. Separate from the DeAlign uncensored checkpoint.
- Model ID
- Tier
- Open weights
- Context
- 512,000 tokens
- Max output
- 128,000 tokens
- Added
- Oct 31
Rank 16 this week: OrcaCyber K3.1 ProorcarouterCapabilities: Reasoning, Tools, Vision, Structured output1M$4.00$20.00
An OrcaRouter model for authorized security research, coding, and tool use. Accepts text and images, with reasoning enabled by default and an option to turn it off. Supports a 1M-token context window, up to 943K output tokens, tool calling, and structured output. Offensive work is permitted only within an authorized security engagement.
- Model ID
- Tier
- Frontier
- Context
- 1,048,064 tokens
- Max output
- 943,718 tokens
- Added
- Nov 5
Rank 17 this week: MiMo V2.6 Flash UncensoredxiaomiCapabilities: Reasoning, Tools, Vision1M$0.50$1.50
A lower-refusal MiMo V2.6 Flash finetune with a 1M-token context window, separate reasoning output, and tool calling. Supports image input and adjustable thinking effort.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 65,536 tokens
- Added
- Sep 25
Rank 18 this week: GLM 5.3 Flash CybersecurityZ.aiCapabilities: Reasoning, Tools, Vision1M$0.15$0.50
GLM 5.3 Flash Cybersecurity is a cybersecurity-focused variant based on the uncensored model, with provider moderation for illegal activities. It supports always-on reasoning, image understanding, tool calling, and a 1,048,576-token context window.
- Model ID
- Tier
- Open weights
- Context
- 1,048,576 tokens
- Max output
- 32,768 tokens
- Added
- Sep 19
Rank 19 this week: DeepSeek V4.1 Flash 1017 AbliteratedDeepSeekCapabilities: Reasoning, Tools1M$0.375$1.50
An abliterated (refusal-removed) version of DeepSeek V4.1 Flash 1017. Text only, with tool calling, optional reasoning and a 1M-token context window. Its behavior may differ from DeepSeek V4.1 Flash 1017. Served by a single host; it may be unavailable while that host is down.
- Model ID
- Tier
- Efficient
- Context
- 1,000,000 tokens
- Max output
- 131,072 tokens
- Added
- Yesterday
Rank 20 this week: OrcaCyber Zero 1.5orcarouterCapabilities: Reasoning, Tools, Structured output999.5K$3.00$7.50
OrcaRouter's frontier cybersecurity model, post-trained for authorized security research: vulnerability research and reproduction, exploit development, penetration testing, and security auditing. Supports a 1M-token context window, optional reasoning, tool calling, and structured output. Offensive work is permitted only within an authorized security engagement.
- Model ID
- Tier
- Frontier
- Context
- 999,488 tokens
- Max output
- 128,000 tokens
- Added
- Oct 30
Reference pricing in USD per 1M tokens, as published by model providers. Tiers describe model metadata, not benchmark rankings.