VibeThinker 3B Heretic is an open-source language model by zkxxxx. Features: 3b LLM, VRAM: 6.2GB, Context: 128K, License: mit, Quantized, Instruction-Based, Code Generating, LLM Explorer Score: 0.33.
Arxiv:2606.16140 Abliterated Base model:quantized:qwen/qwen... Base model:qwen/qwen2.5-coder-... Code Codegen Conversational Decensored En Endpoints compatible Gguf Gpqa Heretic Instruct Instruction-following Math Q4 Quantized Qwen2 Reasoning Region:us Reproducible Safetensors Sharded Tensorflow Uncensored
Model Card on HF 🤗: https://huggingface.co/zkxxxx/VibeThinker-3B-heretic
VibeThinker 3B Heretic Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
VibeThinker 3B Heretic Parameters and Internals
| LLM Name | VibeThinker 3B Heretic |
| Repository 🤗 | https://huggingface.co/zkxxxx/VibeThinker-3B-heretic |
| Base Model(s) | |
| Model Size | 3b |
| Required VRAM | 6.2 GB |
| Updated | 2026-06-25 |
| Maintainer | zkxxxx |
| Model Type | qwen2 |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| GGUF Quantization | Yes |
| Quantization Type | gguf|q4|q4_k |
| Generates Code | Yes |
| Model Architecture | Qwen2ForCausalLM |
| License | mit |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Transformers Version | 5.10.2 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Errors | replace |
Best Alternatives to VibeThinker 3B Heretic
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Coder GRPO 3B | 32K / 6.2 GB | 1209 | 7 |
| Tiny Agent A 3B | 32K / 6.2 GB | 680 | 24 |
| ...2.5 Coder 3B Instruct Bnb 4bit | 32K / 2 GB | 18019 | 7 |
| Qwen2.5 Coder 3B Instruct 4bit | 32K / 1.7 GB | 1621 | 2 |
| Qwen3b Lora Finetune | 32K / 6.2 GB | 0 | 1 |
| Qwen3b Lora Finetune | 32K / 6.2 GB | 0 | 1 |
| Tessa T1 3B | 117K / 6.2 GB | 6 | 5 |
| UIGEN T1.5 3B | 117K / 6.2 GB | 10 | 1 |
| Qwen2.5 Coder 3B Instruct | 32K / 6.2 GB | 451109 | 101 |
| Qwen 2.5 3b Smarteaz 0.01a | 32K / 6.3 GB | 0 | 1 |
Rank the VibeThinker 3B Heretic Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 54529 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a