GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
saidutta69
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
saidutta69
Model Tree
openbmb/MiniCPM5-2B
Input Modalities
Output Modalities
Supported Functionality
Full GGUF ladder included — pick the quant that fits your card:
| Your GPU | Recommended quant | Weights |
|---|---|---|
| RTX 3060 / 4070 / 5070 (12 GB) | Q8_0 | 2.68 GB |
| RTX 4060 / 3070 (8 GB) | Q6_K | 2.07 GB |
| GTX 1660 Super / 2060 / 3050 laptop (6 GB) | Q5_K_M | 1.81 GB |
| CPU-only / Apple Silicon | Q4_K_M | 1.56 GB, fits in system RAM |
Weights only, at this model's 2B native size; add ~1 GB for context. OOM? Drop one quant level. Headroom to spare? Go one up.
Fine-tuning a "helpful" persona on top of RLHF'd refusals fights the base model's training and tends to degrade coherence. Abliteration instead finds and edits the specific weight directions responsible for refusal, leaving the rest of the network (and its capabilities) untouched. See the Heretic repo and the original abliteration writeup for the mechanism.
| Parameter | Value |
|---|---|
| direction_index | per layer |
| attn.o_proj.max_weight | 1.44 |
| attn.o_proj.max_weight_position | 28.24 |
| attn.o_proj.min_weight | 1.16 |
| attn.o_proj.min_weight_distance | 13.17 |
| mlp.down_proj.max_weight | 1.42 |
| mlp.down_proj.max_weight_position | 28.73 |
| mlp.down_proj.min_weight | 1.11 |
| Metric | This model | Original model (openbmb/MiniCPM5-2B) |
|---|---|---|
| KL divergence | 0.0646 | 0 (by definition) |
| Refusals | 4/100 | 99/100 |
Refusals dropped from 99 to 4 out of 100 while preserving MiniCPM5-2B's tool-use, code, and reasoning abilities.
Made with ❤️ by RACER IS OP — follow for more uncensored models
Full quantization set (14 quants + F16) produced with llama.cpp.
| File | Format | Size |
|---|---|---|
MiniCPM5-2B-heretic-F16.gguf | GGUF F16 | 5.04 GB |
MiniCPM5-2B-heretic-Q2_K.gguf | GGUF Q2_K | 1.04 GB |
MiniCPM5-2B-heretic-IQ3_S.gguf | GGUF IQ3_S | 1.19 GB |
MiniCPM5-2B-heretic-Q3_K_S.gguf | GGUF Q3_K_S | 1.19 GB |
MiniCPM5 architecture — loads natively in llama.cpp / Ollama / LM Studio / Jan.
Run llama serve -hf saidutta69/MiniCPM5-2B-heretic to pull the default quant.
bash
# llama.cppllama serve -hf saidutta69/MiniCPM5-2B-heretic
python
# transformersfrom transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "saidutta69/MiniCPM5-2B-heretic"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id,torch_dtype="auto",device_map="auto",)messages = [{"role": "user", "content": "Who are you? Please briefly introduce yourself."}]inputs = tokenizer.apply_chat_template(messages,tokenize=True,add_generation_prompt=True,enable_thinking=False, # set True for Think modereturn_dict=True,return_tensors="pt",).to(model.device)outputs = model.generate(**inputs, max_new_tokens=128)print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))
Also runnable via Ollama, LM Studio, Jan, vLLM, SGLang. For tool/function calling, SGLang is the recommended backend; MiniCPM5 emits XML-style tool calls that SGLang's built-in minicpm5 parser converts to OpenAI-compatible tool_calls.
Refusal suppression is deliberate and works as intended: this model will comply with requests the base model would refuse, including some it shouldn't. There is no safety filtering layered on top. You are responsible for how you deploy it — don't put this behind an unmoderated public-facing endpoint serving third parties. It inherits MiniCPM5-2B's factual limitations and biases; abliteration removes refusal directions, it doesn't add capability or judgment.
Inherits the Apache 2.0 license from the base model.
See the base model card at openbmb/MiniCPM5-2B for the original architecture, training details, requirements, and citation.
| mlp.down_proj.min_weight_distance | 20.52 |
MiniCPM5-2B-heretic-Q3_K_M.gguf |
| GGUF Q3_K_M |
| 1.29 GB |
MiniCPM5-2B-heretic-Q3_K_L.gguf | GGUF Q3_K_L | 1.38 GB |
MiniCPM5-2B-heretic-IQ4_XS.gguf | GGUF IQ4_XS | 1.43 GB |
MiniCPM5-2B-heretic-Q4_K_S.gguf | GGUF Q4_K_S | 1.50 GB |
MiniCPM5-2B-heretic-Q4_0.gguf | GGUF Q4_0 | 1.49 GB |
MiniCPM5-2B-heretic-Q4_1.gguf | GGUF Q4_1 | 1.63 GB |
MiniCPM5-2B-heretic-Q4_K_M.gguf | GGUF Q4_K_M | 1.56 GB |
MiniCPM5-2B-heretic-Q5_K_S.gguf | GGUF Q5_K_S | 1.77 GB |
MiniCPM5-2B-heretic-Q5_K_M.gguf | GGUF Q5_K_M | 1.81 GB |
MiniCPM5-2B-heretic-Q6_K.gguf | GGUF Q6_K | 2.07 GB |
MiniCPM5-2B-heretic-Q8_0.gguf | GGUF Q8_0 | 2.68 GB |