MiniCPM5-1B-Claude-Opus-Fable5-Thinking

<p align="center"> <img src="assets/banner.png" alt="MiniCPM5-1B-Claude-Opus-Fable5-Thinking" width="100%"/> </p>

MiniCPM5-1B-Claude-Opus-Fable5-Thinking

📢 V2.0 is available — We have released an updated model with enhanced tool-calling capabilities. Welcome to try the new version:

• Transformers: MiniCPM5-1B-Claude-Opus-Fable5-V2-Thinking
• GGUF: MiniCPM5-1B-Claude-Opus-Fable5-V2-Thinking-GGUF

GGUF quantizations for local deployment: MiniCPM5-1B-Claude-Opus-Fable5-Thinking-GGUF

中文说明

MiniCPM5-1B-Claude-Opus-Fable5-Thinking is a compact 1B Thinking language model built on openbmb/MiniCPM5-1B. It is further fine-tuned on Fable 5 data to improve coding and instruction-following while keeping MiniCPM5's native Thinking chat template and tool-call format.

For llama.cpp / Ollama / LM Studio deployment, see the GGUF repository.

---

Overview

---

Capabilities

• Coding — code generation, debugging, and software-engineering-style tasks
• Instruction following — more reliable adherence to user prompts and structured constraints
• Thinking mode — chain-of-thought reasoning via the MiniCPM5 chat template
• Tool calling — inherits MiniCPM5's XML tool-call format
• Long context — up to 128K tokens (131,072 tokens per config.json)

---

Quick start

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "GnLOLot/MiniCPM5-1B-Claude-Opus-Fable5-Thinking"

tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    trust_remote_code=True,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

messages = [{"role": "user", "content": "Write a Python function to merge two sorted lists."}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

---

Sampling recommendations

Generation defaults are inherited from MiniCPM5-1B:

---

Limitations

• Thinking outputs — the model may emit reasoning blocks before the final answer; downstream apps can strip them before display
• 1B scale — optimized for lightweight local deployment, not frontier-scale general reasoning

---

Provenance & licensing

Released under Apache-2.0, inherited from MiniCPM5-1B.

Acknowledgements

• Base model: OpenBMB / MiniCPM5-1B
• GGUF conversion: llama.cpp