huihui

deepseek-r1-abliterated

DeepSeek's first generation reasoning models with comparable performance to OpenAI-o1.

691.5K Pulls 55 Tags Updated 12 months ago

qwen3-abliterated

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

tools thinking 0.6b 1.7b 4b 8b 14b 16b 30b 32b 235b

188.6K Pulls 74 Tags Updated 9 months ago

qwen3.5-abliterated

Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.

vision tools thinking 0.8b 2b 4b 9b 27b 35b 122b

177.3K Pulls 58 Tags Updated 1 month ago

gemma-4-abliterated

Gemma 4 models are designed to deliver frontier-level performance at each size. They are well-suited for reasoning, agentic workflows, coding, and multimodal understanding.

vision tools thinking audio e2b e4b 26b 31b 48b

170.6K Pulls 29 Tags Updated 1 month ago

gemma3-abliterated

The current, most capable model that runs on a single GPU.

vision 270m 1b 4b 12b 27b

85.3K Pulls 16 Tags Updated 9 months ago

qwen3-vl-abliterated

The most powerful vision-language model in the Qwen3 model family to date.

vision tools thinking 2b 4b 8b 30b 32b

84.3K Pulls 54 Tags Updated 6 months ago

qwq-abliterated

QwQ is an experimental research model focused on advancing AI reasoning capabilities.

tools 32b

71K Pulls 17 Tags Updated 1 year ago

glm-4.7-flash-abliterated

As the strongest model in the 30B class, GLM-4.7-Flash offers a new option for lightweight deployment that balances performance and efficiency.

tools thinking

54.9K Pulls 5 Tags Updated 3 months ago

qwen2.5-abliterate

Qwen2 is a new series of large language models from Alibaba group

tools 0.5b 1.5b 3b 7b 14b 32b 72b

48.2K Pulls 46 Tags Updated 1 year ago

exaone3.5-abliterated

EXAONE 3.5 is a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.

2.4b 7.8b 32b

44.3K Pulls 13 Tags Updated 1 year ago

llama3.2-abliterate

Meta's Llama 3.2 goes small with 1B and 3B models.

tools 1b 3b

42K Pulls 11 Tags Updated 1 year ago

gpt-oss-abliterated

OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.

tools thinking 20b 120b

41.3K Pulls 16 Tags Updated 7 months ago

qwen2.5-1m-abliterated

Qwen2.5-1M is the long-context version of the Qwen2.5 series models, supporting a context length of up to 1M tokens.

tools 7b 14b

40.7K Pulls 11 Tags Updated 1 year ago

dolphin3-abliterated

Dolphin 3.0 Llama 3.1 8B 🐬 is the next generation of the Dolphin series of instruct-tuned models designed to be the ultimate general purpose local model, enabling coding, math, agentic, function calling, and general use cases.

tools 8b

39.7K Pulls 5 Tags Updated 1 year ago

llama3.3-abliterated

New state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model.

tools 70b

37.5K Pulls 10 Tags Updated 1 year ago

Qwen3.6-abliterated

Qwen3.6 delivers substantial upgrades in agentic coding and thinking preservation than previous Qwen models.

vision tools thinking 27b 35b

26.1K Pulls 17 Tags Updated 1 month ago

qwen2.5-coder-abliterate

The latest series of Code-Specific Qwen models, with significant improvements in code generation, code reasoning, and code fixing.

tools 0.5b 1.5b 3b 7b 14b 32b

20.9K Pulls 31 Tags Updated 1 year ago

gemma3n-abliterated

This is an uncensored version of google/gemma-3n created with abliteration

20.6K Pulls 2 Tags Updated 10 months ago

huihui-moe-abliterated

Huihui-MoE-abliterated is a **Mixture of Experts (MoE)** language model developed by **huihui.ai**

tools thinking 1.5b 5b 12b 23b 24b 46b 57b 60b

13.2K Pulls 40 Tags Updated 8 months ago

phi4-abliterated

Phi-4 is a 14B parameter, state-of-the-art open model from Microsoft.

14b

12.6K Pulls 5 Tags Updated 1 year ago

qwen3-coder-abliterated

Qwen3-Coder featuring the following key enhancements: Significant Performance, Long-context Capabilities, Agentic Coding.

30b 480b

12.5K Pulls 9 Tags Updated 9 months ago

kimi-k2

This is not the ablation version. Kimi-K2-Instruct is a state-of-the-art mixture-of-experts (MoE) language model with 32 billion activated parameters and 1 trillion total parameters.

tools 1026b

11.5K Pulls 4 Tags Updated 10 months ago

hy-mt1.5-abliterated

Hunyuan Translation Model Version 1.5 includes a 1.8B translation model, HY-MT1.5-1.8B, and a 7B translation mode

1.8b 7b

11.1K Pulls 16 Tags Updated 4 months ago

mistral-small-abliterated

Mistral Small 3 sets a new benchmark in the “small” Large Language Models category below 70B.

tools 24b

10.5K Pulls 10 Tags Updated 1 year ago

qwen2.5-vl-abliterated

Flagship vision-language model of Qwen and also a significant leap from the previous Qwen2-VL.

vision 3b 7b 32b

8,241 Pulls 16 Tags Updated 6 months ago

baronllm-abliterated

This is an uncensored version of AlicanKiraz0/BaronLLM_Offensive_Security_LLM_Q6_K_GGUF created with abliteration

tools 8b

7,880 Pulls 4 Tags Updated 11 months ago

hunyuan-mt-abliterated

The Hunyuan Translation Model comprises a translation model, Hunyuan-MT-7B, and an ensemble model, Hunyuan-MT-Chimera.

7b

7,698 Pulls 9 Tags Updated 8 months ago

deepseek-v3

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

7,181 Pulls 2 Tags Updated 1 year ago

openthinker-abliterated

A fully open-source family of reasoning models built using a dataset derived by distilling DeepSeek-R1.

7b 32b

6,572 Pulls 9 Tags Updated 1 year ago

magistral-abliterated

This is an uncensored version of mistralai/magistral-Small-2506 created with abliteration

thinking 24b

6,493 Pulls 6 Tags Updated 11 months ago

granite3.2-vision-abliterated

A compact and efficient vision-language model, specifically designed for visual document understanding, enabling automated content extraction from tables, charts, infographics, plots, diagrams, and more.

vision tools 2b

6,368 Pulls 5 Tags Updated 1 year ago

phi4-mini-abliterated

Phi-4-mini brings significant enhancements in multilingual support, reasoning, and mathematics, and now, the long-awaited function calling feature is finally supported.

tools 3.8b

6,116 Pulls 5 Tags Updated 1 year ago

dolphin3-r1-abliterated

Dolphin's first generation reasoning models.

24b

5,987 Pulls 10 Tags Updated 1 year ago

qwen3-coder-next-abliterated

Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.

tools

5,894 Pulls 4 Tags Updated 3 months ago

qwen3-next-abliterated

The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.

tools thinking 80b

4,971 Pulls 10 Tags Updated 5 months ago

deephermes3-abliterated

DeepHermes 3 Preview is the latest version of our flagship Hermes series of LLMs by Nous Research, and one of the first models in the world to unify Reasoning (long chains of thought that improve answer accuracy) and normal LLM response modes into one mod

8b

3,839 Pulls 6 Tags Updated 1 year ago

deepseek-v3-abliterated

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

671b

3,839 Pulls 5 Tags Updated 1 year ago

devstral-abliterated

This is an uncensored version of mistralai/Devstral-Small-2505 created with abliteration

tools 24b

3,775 Pulls 6 Tags Updated 12 months ago

lfm2.5-abliterated

LFM2.5 is a new family of hybrid models designed for on-device deployment.

tools 1.2b

3,180 Pulls 10 Tags Updated 3 months ago

homunculus-abliterated

This is an uncensored version of arcee-ai/Homunculus created with abliteration

thinking 12b

3,107 Pulls 5 Tags Updated 11 months ago

jan-nano-abliterated

This is an uncensored version of Menlo/Jan-nano created with abliteration

tools thinking 4b

3,027 Pulls 10 Tags Updated 10 months ago

seed-coder-abliterate

This is an uncensored version of ByteDance-Seed/Seed-Coder-8B-Instruct created with abliteration

3,016 Pulls 5 Tags Updated 1 year ago

foundation-sec-abliterated

Foundation-Sec-8B-abliterated is an uncensored version fine-tuned based on fdtn-ai/Foundation-Sec-8B. Foundation-Sec-8B is an open-weight, 8-billion-parameter foundational language model designed specifically for cybersecurity applications.

8b

3,007 Pulls 5 Tags Updated 1 year ago

phi4-reasoning-abliterated

Phi 4 mini reasoning is a lightweight open model that balances efficiency with advanced reasoning ability.

3.8b

2,999 Pulls 4 Tags Updated 1 year ago

am-thinking-abliterate

This is an uncensored version of a-m-team/AM-Thinking-v1 created with abliteration

tools 32b

2,977 Pulls 6 Tags Updated 1 year ago

granite4.1-abliterated

Granite 4.1 models have gone through an improved post-training pipeline, including supervised finetuning and reinforcement learning alignment, resulting in enhanced tool calling, instruction following, and chat capabilities.

tools 3b 8b 30b

2,891 Pulls 12 Tags Updated 2 weeks ago

qwenlong-abliterated

This is an uncensored version of Tongyi-Zhiwen/QwenLong-L1-32B created with abliteration

thinking 32b

2,879 Pulls 5 Tags Updated 11 months ago

nemotron-abliterated

Llama-3.1-Nemotron-70B-Instruct is a large language model customized by NVIDIA to improve the helpfulness of LLM generated responses to user queries.

tools 70b

2,845 Pulls 6 Tags Updated 1 year ago

tongyi-deepresearch-abliterated

tongyi DeepResearch, an agentic large language model featuring 30 billion total parameters, with only 3 billion activated per token.

tools thinking 30b

2,792 Pulls 5 Tags Updated 8 months ago

Hermes-3-Llama-3.2-abliterated

Hermes 3 3B is a small but mighty new addition to the Hermes series of LLMs by Nous Research, and is Nous's first fine-tune in this parameter class.

tools 3b

2,625 Pulls 5 Tags Updated 1 year ago

granite3.2-abliterated

Granite-3.2 is a family of long-context AI models from IBM Granite fine-tuned for thinking capabilities.

tools 2b 8b

2,578 Pulls 11 Tags Updated 1 year ago

aya-expanse-abliterated

Cohere For AI's language models trained to perform well across 23 different languages.

tools 8b 32b

2,498 Pulls 9 Tags Updated 1 year ago

tess-r1-abliterated

Tess-R1 is designed with test-time compute in mind, and has the capabilities to produce a Chain-of-Thought (CoT) reasoning before producing the final output.

70b

2,440 Pulls 9 Tags Updated 1 year ago

deepseek-r1-Fusion

DeepSeek-R1-Distill-Qwen-Coder-32B-Fusion-9010 is a mixed model that combines the strengths of two powerful DeepSeek-R1-Distill-Qwen-based models: huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated and huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated.

32b

2,435 Pulls 6 Tags Updated 1 year ago

arcee-blitz-abliterated

Arcee-Blitz (24B) is a new Mistral-based 24B model distilled from DeepSeek, designed to be both fast and efficient. We view it as a practical “workhorse” model that can tackle a range of tasks without the overhead of larger architectures.

24b

2,428 Pulls 5 Tags Updated 1 year ago

deepseekr1-qwq-skyt1-fusion

`DeepSeekR1-QwQ-SkyT1-32B-Fusion` is a mixed model that combines the strengths of three powerful Qwen-based models: huihui-ai/DeepSeek-R1-Distill-Qwen-32B-abliterated, huihui-ai/QwQ-32B-Preview-abliterated and huihui-ai/Sky-T1-32B-Preview-abliterated

32b

2,385 Pulls 14 Tags Updated 1 year ago

acereason-nemotron-abliterated

This is an uncensored version of nvidia/AceReason-Nemotron created with abliteration Edit

7b 14b

2,377 Pulls 9 Tags Updated 12 months ago

qwenlong-l1.5-abliterated

QwenLong-L1.5, a long-context reasoning model built upon Qwen3-30B-A3B-Thinking, augmented with memory mechanisms to process tasks far beyond its physical context window.

tools thinking 30b

2,316 Pulls 8 Tags Updated 5 months ago

mirothinker1-abliterated

MiroThinker v1.0 is an open-source research agent designed to advance tool-augmented reasoning and information-seeking capabilities.

tools thinking 8b 30b 72b

2,289 Pulls 18 Tags Updated 6 months ago

fara-abliterated

Fara-7B is Microsoft's first agentic small language model (SLM) designed specifically for computer use.

vision 7b

2,266 Pulls 5 Tags Updated 5 months ago

exaone-deep-abliterated

This is an uncensored version of LGAI-EXAONE/EXAONE-Deep-2.4B created with abliteration

2.4b 7.8b

1,985 Pulls 9 Tags Updated 11 months ago

skywork-o1-abliterated

The Skywork o1 Open model series, developed by the Skywork team at Kunlun Inc. This groundbreaking release introduces a series of models that incorporate o1-like slow thinking and reasoning capabilities.

tools 8b

1,920 Pulls 5 Tags Updated 1 year ago

deepscaler-abliterated

A fine-tuned version of Deepseek-R1-Distilled-Qwen-1.5B that surpasses the performance of OpenAI’s o1-preview with just 1.5B parameters on popular math evaluations.

1.5b

1,890 Pulls 3 Tags Updated 1 year ago

orchestrator-abliterated

Orchestrator-8B is a state-of-the-art 8B parameter orchestration model designed to solve complex, multi-turn agentic tasks by coordinating a diverse set of expert models and tools.

tools thinking 8b

1,769 Pulls 5 Tags Updated 5 months ago

falcon3-abliterated

A family of efficient AI models under 10B parameters performant in science, math, and coding through innovative training techniques.

1b 3b 7b 10b

1,711 Pulls 21 Tags Updated 1 year ago

llama3.3-abliterated-ft

The fine-tuned version of huihui_ai/llama3.3-abliterated

tools 70b

1,702 Pulls 4 Tags Updated 1 year ago

huihui-4

Huihui4-8B-A4B is a lightweight MoE (Mixture of Experts) conversational model optimized from Google's gemma-4-26B-A4B-it architecture

vision tools thinking 8b

1,625 Pulls 4 Tags Updated 1 month ago

tinyr1-abliterated

Tiny-R1-32B-Preview, which outperforms the 70B model Deepseek-R1-Distill-Llama-70B and nearly matches the full R1 model in math.

tools 32b

1,588 Pulls 6 Tags Updated 1 year ago

kanana-nano-abliterated

Kanana, a series of bilingual language models (developed by Kakao) that demonstrate exceeding performance in Korean and competitive performance in English.

2.1b

1,470 Pulls 5 Tags Updated 1 year ago

marco-o1-abliterated

An open large reasoning model for real-world solutions by the Alibaba International Digital Commerce Group (AIDC-AI).

7b

1,402 Pulls 5 Tags Updated 1 year ago

deepseek-v3-pruned

DeepSeek-V3-Pruned-Coder-411B is a pruned version of the DeepSeek-V3 reduced from 256 experts to 160 experts, The pruned model is mainly used for code generation.

411b

1,377 Pulls 5 Tags Updated 1 year ago

smallthinker-abliterated

A new small reasoning model fine-tuned from the Qwen 2.5 3B Instruct model.

3b

1,374 Pulls 5 Tags Updated 1 year ago

granite3.1-dense-abliterated

The IBM Granite 2B and 8B models are text-only dense LLMs trained on over 12 trillion tokens of data, demonstrated significant improvements over their predecessors in performance and speed in IBM’s initial testing.

tools 2b 8b

1,290 Pulls 11 Tags Updated 1 year ago

qwq-fusion

qwq-fusion is a mixed model that combines the strengths of two powerful Qwen-based models: huihui-ai/QwQ-32B-Preview-abliterated and huihui-ai/Qwen2.5-Coder-32B-Instruct-abliterated.

tools 32b

1,203 Pulls 14 Tags Updated 1 year ago

internlm3-abliterated

InternLM3 has open-sourced an 8-billion parameter instruction model, InternLM3-8B-Instruct, designed for general-purpose usage and advanced reasoning.

8b

1,103 Pulls 5 Tags Updated 1 year ago

command-r7b-abliterated

he smallest model in Cohere's R series delivers top-tier speed, efficiency, and quality to build powerful AI applications on commodity GPUs and edge devices.

tools 7b

1,063 Pulls 5 Tags Updated 1 year ago

perplexity-ai-r1-abliterated

A version of the DeepSeek-R1 model that has been post trained to provide unbiased, accurate, and factual information by Perplexity.

70b

1,019 Pulls 10 Tags Updated 1 year ago

huihui-moe

Huihui-MoE is a Mixture of Experts (MoE) language model developed by huihui.ai

tools thinking 1.2b 23b

1,009 Pulls 10 Tags Updated 11 months ago

nemotron-v1-abliterated

Llama Nemotron, Open, Production-ready Enterprise Models

tools 8b

1,007 Pulls 6 Tags Updated 1 year ago

openhands-lm-abliterated

OpenHands LM is built on the foundation of Qwen Coder 2.5 Instruct 32B, leveraging its powerful base capabilities for coding tasks.

tools 32b

896 Pulls 7 Tags Updated 1 year ago

microthinker

MicroThinker is an experimental research model focused on advancing AI reasoning capabilities.

tools 1b 3b 8b

840 Pulls 12 Tags Updated 1 year ago

tulu3-abliterate

Tülu 3 is a leading instruction following model family, offering fully open-source data, code, and recipes by the The Allen Institute for AI.

8b 70b

832 Pulls 9 Tags Updated 1 year ago

Qwen3-Coder

This is not the ablation version. Qwen3-Coder featuring the following key enhancements: Significant Performance, Long-context Capabilities, Agentic Coding.

tools thinking 480b

750 Pulls 4 Tags Updated 10 months ago

glm-4.7-abliterated

Advancing the Coding Capability

tools thinking

738 Pulls 1 Tag Updated 3 months ago

kimi-k2-abliterated

A state-of-the-art mixture-of-experts (MoE) language model. Kimi K2-Instruct-0905 demonstrates significant improvements in performance on public benchmarks and real-world coding agent tasks.

tools 1026b

689 Pulls 6 Tags Updated 5 months ago

perplexity-ai-r1

A version of the DeepSeek-R1 model that has been post trained to provide unbiased, accurate, and factual information by Perplexity.

518 Pulls 2 Tags Updated 1 year ago

glm4.6-abliterated

Advanced agentic, reasoning and coding capabilities.

tools thinking 357b

488 Pulls 4 Tags Updated 5 months ago

uwu-abliterated

UwU is an experimental research model focused on advancing AI reasoning capabilities.

7b

471 Pulls 5 Tags Updated 1 year ago

devstral-2-abliterated

123B model that excels at using tools to explore codebases, editing multiple files and power software engineering agents.

tools 123b

415 Pulls 5 Tags Updated 4 months ago

s1.1-abliterated

This model is a successor of s1-32B with slightly better performance.

tools 32b

375 Pulls 4 Tags Updated 1 year ago

Lucie-abliterated

Lucie-7B is a pretrained 7B parameter causal language model built by LINAGORA and OpenLLM-France.

7b

357 Pulls 5 Tags Updated 1 year ago

s1-abliterated

s1 is a reasoning model finetuned from Qwen2.5-32B-Instruct on just 1,000 examples. It matches o1-preview & exhibits test-time scaling via budget forcing.

tools 32b

356 Pulls 9 Tags Updated 1 year ago

huihui3.5

Huihui 3.5 is a family of models created by remixing and merging multiple Qwen3.5 models of the same size.

vision tools thinking 67b

323 Pulls 1 Tag Updated 1 month ago

deepseek-r1

DeepSeek's first-generation of reasoning models with comparable performance to OpenAI-o1, including six dense models distilled from DeepSeek-R1 based on Llama and Qwen.

thinking 671b

257 Pulls 12 Tags Updated 11 months ago

fluentlylm-prinum-abliterated

fluently-lm/FluentlyLM-Prinum

tools 32b

255 Pulls 5 Tags Updated 1 year ago

deepseek-v3.1

This is not the ablation version. DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode.

tools thinking 671b

216 Pulls 3 Tags Updated 8 months ago

skyt1-abliterated

Sky-T1-32B-Preview is an experimental research model focused on advancing AI reasoning capabilities

tools 32b

214 Pulls 5 Tags Updated 1 year ago

qwen2.5-censortune

CensorTune with Supervised Fine-Tuning (SFT) to fine-tune the Qwen2.5-Instruct model on 622 harmful instructions in a single fine-tuning iteration, achieving rejection of these instructions and a zero-pass rate for 320

tools 0.5b 1.5b 3b

214 Pulls 15 Tags Updated 1 year ago

megrez-abliterated

Megrez-3B aims to provide a fast inference, compact, and powerful edge-side intelligent solution through software-hardware co-design.

tools 7b

179 Pulls 6 Tags Updated 1 year ago

deepseek-r1-pruned

DeepSeek-R1-Pruned-Coder-411B is a pruned version of the DeepSeek-R1 reduced from 256 experts to 160 experts, The pruned model is mainly used for code generation.

411b

88 Pulls 3 Tags Updated 1 year ago

[email protected]

deepseek-r1-abliterated

qwen3-abliterated

qwen3.5-abliterated

gemma-4-abliterated

gemma3-abliterated

qwen3-vl-abliterated

qwq-abliterated

glm-4.7-flash-abliterated

qwen2.5-abliterate

exaone3.5-abliterated

llama3.2-abliterate

gpt-oss-abliterated

qwen2.5-1m-abliterated

dolphin3-abliterated

llama3.3-abliterated

Qwen3.6-abliterated

qwen2.5-coder-abliterate

gemma3n-abliterated

huihui-moe-abliterated

phi4-abliterated

qwen3-coder-abliterated

kimi-k2

hy-mt1.5-abliterated

mistral-small-abliterated

qwen2.5-vl-abliterated

baronllm-abliterated

hunyuan-mt-abliterated

deepseek-v3

openthinker-abliterated

magistral-abliterated

granite3.2-vision-abliterated

phi4-mini-abliterated

dolphin3-r1-abliterated

qwen3-coder-next-abliterated

qwen3-next-abliterated

deephermes3-abliterated

deepseek-v3-abliterated

devstral-abliterated

lfm2.5-abliterated

homunculus-abliterated

jan-nano-abliterated

seed-coder-abliterate

foundation-sec-abliterated

phi4-reasoning-abliterated

am-thinking-abliterate

granite4.1-abliterated

qwenlong-abliterated

nemotron-abliterated

tongyi-deepresearch-abliterated

Hermes-3-Llama-3.2-abliterated

granite3.2-abliterated

aya-expanse-abliterated

tess-r1-abliterated

deepseek-r1-Fusion

arcee-blitz-abliterated

deepseekr1-qwq-skyt1-fusion

acereason-nemotron-abliterated

qwenlong-l1.5-abliterated

mirothinker1-abliterated

fara-abliterated

exaone-deep-abliterated

skywork-o1-abliterated

deepscaler-abliterated

orchestrator-abliterated

falcon3-abliterated

llama3.3-abliterated-ft

huihui-4

tinyr1-abliterated

kanana-nano-abliterated

marco-o1-abliterated

deepseek-v3-pruned

smallthinker-abliterated

granite3.1-dense-abliterated

qwq-fusion

internlm3-abliterated

command-r7b-abliterated

perplexity-ai-r1-abliterated

huihui-moe

nemotron-v1-abliterated