# Open Uncensored models

48 uncensored and abliterated text models and 17 uncensored image models behind one OpenAI compatible API at `https://openuncensored.com/api/v1`.
Prices are USD per million tokens at the cheapest provider, and that is exactly what you pay: no markup, no commission
(cached prompt tokens at the provider's cached price). Sorted by OpenUncensored Score rank.

Machine-readable: `GET https://openuncensored.com/api/v1/models` (no key needed) or `https://openuncensored.com/api/rankings.json`.

| id | name | input $/M | output $/M | context | providers | OU Score | speed (p50 first token · tok/s) |
|---|---|---|---|---|---|---|---|
| `deepseek-v4.1-flash-1017-uncensored` | DeepSeek V4.1 Flash 1017 Uncensored | $0.330 | $1.32 | 1049K | venice, nanogpt | #1 (69.8) | 758ms · 264 tok/s |
| `deepseek-v4.1-flash-1017-uncensored-thinking` | DeepSeek V4.1 Flash 1017 Uncensored Thinking | $0.330 | $1.32 | 1049K | nanogpt | #1 (69.8) | 7.0s · 197 tok/s |
| `glm-5.3-flash-uncensored` | GLM 5.3 Flash Uncensored | $0.200 | $0.800 | 1049K | nanogpt | #2 (69.7) | 2.6s · 111 tok/s |
| `qwen3.8-27b-uncensored-fast` | Qwen 3.8 27B Uncensored Fast by Heabsy | $1.00 | $3.00 | 524K | nanogpt | #3 (66.8) | 594ms · 252 tok/s (test) |
| `deepseek-v4.1-flash-0910-uncensored` | DeepSeek V4.1 Flash 0910 Uncensored | $0.330 | $1.32 | 524K | nanogpt | #4 (66.1) | 1.6s · 7 tok/s |
| `deepseek-v4.1-flash-0910-uncensored-thinking` | DeepSeek V4.1 Flash 0910 Uncensored Thinking | $0.330 | $1.32 | 524K | nanogpt | #4 (66.1) | – |
| `deepseek-v4.1-flash-uncensored` | DeepSeek V4.1 Flash Uncensored | $0.330 | $1.32 | 1049K | venice, nanogpt | #5 (64.3) | 912ms · 239 tok/s |
| `deepseek-v4.1-flash-uncensored-thinking` | DeepSeek V4.1 Flash Uncensored Thinking | $0.330 | $1.32 | 1049K | nanogpt | #5 (64.3) | – |
| `abliterated-large-flash` | Abliterated Large Flash | $1.00 | $2.00 | 1049K | abliteration-ai, venice, nanogpt | #6 (64.0) | 1.0s · 159 tok/s |
| `mimo-v2.6-flash-uncensored` | MiMo V2.6 Flash Uncensored | $0.500 | $1.50 | 1049K | nanogpt | #7 (63.5) | 1.1s · 93 tok/s (test) |
| `mimo-v2.6-flash-uncensored-thinking` | MiMo V2.6 Flash Uncensored Thinking | $0.500 | $1.50 | 1049K | nanogpt | #7 (63.5) | – |
| `glm-5.3-flash-apex-abliterated` | Apex Flash 1 Abliterated | $0.150 | $0.500 | 1049K | nanogpt | #8 (63.2) | 6.4s · 11 tok/s |
| `glm-5.3-uncensored` | GLM 5.3 Uncensored | $1.25 | $2.25 | 1049K | nanogpt | #9 (63.0) | 1.7s · 32 tok/s (test) |
| `mimo-v2.6-flash-abliterated` | MiMo V2.6 Flash Abliterated | $0.500 | $1.50 | 1049K | nanogpt | #10 (62.6) | 1.2s · 83 tok/s (test) |
| `abliterated-large-v2` | Abliterated Large v2 | $3.00 | $5.00 | 1049K | abliteration-ai, venice, nanogpt | #11 (62.0) | 4.2s · 110 tok/s |
| `qwen3.8-27b-uncensored` | Phala: Qwen3.8 27B Uncensored (Aggressive) | $0.300 | $1.50 | 1049K | redpill, nanogpt | #12 (62.0) | 1.9s · 16 tok/s |
| `qwen3.8-27b-uncensored-thinking` | Qwen 3.8 27B Uncensored Thinking | $1.00 | $3.00 | 1049K | nanogpt | #12 (62.0) | – |
| `gemma-4-31b-gembrain-uncensored-heretic` | Gemma 4 31B Gembrain Uncensored Heretic | $0.306 | $0.306 | 262K | nanogpt | #13 (58.5) | 2.2s · 11 tok/s (test) |
| `qwen3.5-27b-claude-4.6-opus-reasoning-distilled-derestricted` | Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted | $0.306 | $0.306 | 262K | nanogpt | #14 (57.8) | 2.8s · 28 tok/s (test) |
| `qwen3.5-27b-queen-derestricted` | Qwen3.5 27B Queen Derestricted | $0.306 | $0.306 | 262K | nanogpt | #15 (57.3) | 3.0s · 32 tok/s (test) |
| `qwen3.5-27b-bluestar-v3-derestricted` | Qwen3.5 27B BlueStar v3 Derestricted | $0.306 | $0.306 | 262K | nanogpt | #16 (56.4) | 7.7s · 24 tok/s (test) |
| `qwen3.5-27b-blossom-v6.4-derestricted` | Qwen3.5 27B Blossom V6.4 Derestricted | $0.306 | $0.306 | 262K | nanogpt | #17 (56.3) | 3.4s · 24 tok/s (test) |
| `gemma-4-uncensored` | Gemma 4 Uncensored | $0.163 | $0.500 | 256K | venice | #18 (54.6) | 450ms · 53 tok/s |
| `gemma-4-26b-a4b-uncensored` | Gemma 4 26B A4B Uncensored TEE | $0.150 | $0.700 | 262K | nanogpt, redpill | #19 (52.8) | 1.6s · 44 tok/s (test) |
| `olafangensan-glm-4.7-flash-heretic` | GLM 4.7 Flash Heretic | $0.070 | $0.400 | 200K | venice | #20 (52.7) | 501ms · 90 tok/s (test) |
| `glm-4.6-derestricted-v5` | GLM 4.6 Derestricted v5 | $0.400 | $1.50 | 131K | nanogpt | #21 (50.2) | 2.3s · 29 tok/s (test) |
| `qwen3.5-27b-claude-4.6-opus-reasoning-distilled-derestricted-lite` | Qwen3.5 27B Claude 4.6 Opus Reasoning Distilled Derestricted Lite | $0.306 | $0.306 | 262K | nanogpt | #22 (49.8) | 9.5s · 26 tok/s (test) |
| `venice-uncensored` | Venice Uncensored 1.2 | $0.200 | $0.900 | 128K | venice, nanogpt | #23 (49.3) | 1.6s · 8 tok/s |
| `venice-uncensored-role-play` | Venice Role Play Uncensored | $0.500 | $2.00 | 128K | venice | #24 (48.6) | 548ms · 80 tok/s (test) |
| `qwen2.5-32b-instruct-abliterated` | Qwen 2.5 32B Abliterated | $0.700 | $0.700 | 33K | nanogpt | #25 (44.6) | 3.6s · 19 tok/s (test) |
| `llama-3.3-70b-instruct-abliterated` | Llama 3.3 70B Instruct abliterated | $0.700 | $0.700 | 33K | nanogpt | #26 (41.7) | 2.5s · 19 tok/s (test) |
| `deepseek-r1-distill-llama-70b-abliterated` | DeepSeek R1 Llama 70B Abliterated | $0.700 | $0.700 | 16K | nanogpt | #27 (40.4) | 6.7s · 18 tok/s (test) |
| `deepseek-r1-distill-qwen-32b-abliterated` | DeepSeek R1 Qwen Abliterated | $1.40 | $1.40 | 16K | nanogpt | #28 (40.0) | 3.7s · 20 tok/s (test) |
| `neuraldaredevil-8b-abliterated` | Neural Daredevil 8B abliterated | $0.440 | $0.440 | 8K | nanogpt | #29 (35.8) | 1.8s · 32 tok/s (test) |
| `meta-llama-3-70b-instruct-abliterated-v3.5` | Llama 3 70B abliterated | $0.700 | $0.700 | 8K | nanogpt | #30 (34.4) | 2.9s · 17 tok/s (test) |
| `aion-rp-llama-3.1-8b` | Llama 3.1 8b (uncensored) | $0.800 | $1.60 | 33K | nanogpt | #31 (33.6) | 1.8s · 36 tok/s (test) |
| `abliterated-model` | Abliterated Model | $1.00 | $3.00 | 262K | abliteration-ai, nanogpt | unranked (68.4) | 1.0s · 222 tok/s (test) |
| `aion-labs-aion-3-0` | Aion 3.0 | $3.75 | $7.50 | 128K | venice | unranked (44.3) | 859ms · 50 tok/s (test) |
| `aion-labs-aion-3-0-mini` | Aion 3.0 Mini | $0.875 | $1.75 | 128K | venice | unranked (56.1) | 758ms · 64 tok/s (test) |
| `aion-labs-aion-3-5` | Aion 3.5 | $3.75 | $7.50 | 262K | venice | unranked (63.6) | 1.6s · 56 tok/s (test) |
| `aion-labs-aion-3-5-mini` | Aion 3.5 Mini | $0.875 | $1.75 | 262K | venice | unranked (78.6) | 1.9s · 65 tok/s (test) |
| `cyberheabsy` | CyberHeabsy | $1.50 | $4.50 | 1049K | nanogpt | unranked (59.9) | 718ms · 132 tok/s (test) |
| `deepseek-v4.1-flash-1017-heretic` | DeepSeek V4.1 Flash Heretic | $0.198 | $0.792 | 512K | nanogpt | unranked (72.7) | 1.3s · 85 tok/s (test) |
| `deepseek-v4.1-flash-1017-heretic-thinking` | DeepSeek V4.1 Flash Heretic Thinking | $0.198 | $0.792 | 512K | nanogpt | unranked (72.7) | – |
| `gemma-4-26b-a4b-it-chimeraxheretic` | Gemma 4 26B A4B Chimera X Heretic | $0.120 | $0.380 | 262K | nanogpt | unranked (61.9) | 4.4s · 75 tok/s |
| `gemma-4-31b-cognitive-unshackled` | Gemma 4 31B Cognitive Unshackled | $0.306 | $0.306 | 262K | nanogpt | unranked (59.6) | 2.7s · 19 tok/s (test) |
| `gemma-4-31b-sdft-heretic-rp` | Gemma 4 31B SDFT Heretic RP | $0.306 | $0.306 | 262K | nanogpt | unranked (61.3) | 2.1s · 29 tok/s (test) |
| `grayline-qwen3-8b` | Grayline Qwen3 8B | $0.300 | $0.300 | 33K | nanogpt | not measured | – |

Speed is measured on real API traffic over the last 7 days (`GET https://openuncensored.com/api/v1/stats/latency`) once a model has 5 successful
requests; until then it is our speed test (5 streamed requests through the provider we measure on), marked (test).

Use the `id` as `model` in chat completions. Model detail: `https://openuncensored.com/models/<id>.md`.

## Image models

17 uncensored image models for `POST https://openuncensored.com/api/v1/images/generations`. The price per image is what is charged to your credits.
Machine-readable: `GET https://openuncensored.com/api/v1/images/models` (no key needed).

| id | name | per image | providers | notes |
|---|---|---|---|---|
| `z-image-turbo` | Z-Image Turbo | $0.0100 | venice, nanogpt | Fast photorealistic model from Alibaba Tongyi (Apache 2.0). Good default. |
| `lustify-v8` | Lustify v8 | $0.0100 | venice | SDXL fine-tune for explicit adult photography. Venice tags it most uncensored. |
| `lustify-v7` | Lustify v7 | $0.0100 | venice | Earlier Lustify release, explicit adult photography. |
| `lustify-sdxl` | Lustify SDXL | $0.0100 | venice | The original Lustify SDXL checkpoint. |
| `chroma` | Chroma | $0.0100 | venice, nanogpt | Uncensored FLUX.1-schnell derivative (Apache 2.0), strong at prompt following. |
| `qwen-image` | Qwen Image | $0.0300 | venice | Alibaba's Qwen-Image (Apache 2.0). Best text rendering inside images. |
| `qwen-image-3-uncensored` | Qwen Image 3 Uncensored | $0.0400 | nanogpt | Uncensored Qwen Image 3, high detail at 1K and 2K. |
| `hidream-o1-dev` | HiDream O1 Dev | $0.0200 | nanogpt | HiDream's fast 2K model, allows adult content. |
| `juggernaut-xl` | Juggernaut XL | $0.0130 | civitai | The most downloaded SDXL photoreal checkpoint on Civitai. Runs on Civitai. |
| `cyberrealistic-xl` | CyberRealistic XL | $0.0150 | civitai | Photoreal SDXL checkpoint, strong on people. Runs on Civitai. |
| `pornmaster` | PornMaster | $0.0130 | civitai | SDXL checkpoint for explicit adult photography. Runs on Civitai. |
| `pony-diffusion-v6` | Pony Diffusion V6 XL | $0.0130 | civitai | The original Pony model for stylised and adult art. Start prompts with "score_9, score_8_up, score_7_up". Runs on Civitai. |
| `cyberrealistic-pony` | CyberRealistic Pony | $0.0150 | civitai | Photoreal Pony fine-tune for adult content. Use Pony score tags. Runs on Civitai. |
| `pony-realism` | Pony Realism | $0.0150 | civitai | Realistic Pony fine-tune for adult content. Use Pony score tags. Runs on Civitai. |
| `wai-illustrious` | WAI Illustrious | $0.0130 | civitai | The most popular Illustrious anime checkpoint. Start prompts with "masterpiece, best quality". Runs on Civitai. |
| `hassaku-xl` | Hassaku XL | $0.0130 | civitai | Illustrious based anime checkpoint for adult art. Runs on Civitai. |
| `noobai-xl` | NoobAI XL | $0.0130 | civitai | Illustrious derived anime base model trained on Danbooru and e621 tags. Runs on Civitai. |