GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Rumiii
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Rumiii
Model Tree
Qwen/Qwen3-4B-Instruct-2507
Input Modalities
Output Modalities
Supported Functionality
| Parent (teacher) model | Kimi K2 |
| Student model | Qwen3-4B-Instruct-2507 |
| Dataset | Agent-Ark/Toucan-1.5M (Kimi-K2 configuration) |
| Method | Supervised fine-tuning on teacher trajectories |
Toucan-1.5M contains over 1.5 million agentic trajectories synthesized from 495 real-world Model Context Protocol (MCP) servers, spanning more than 2,000 tools. It includes single-turn and multi-turn interactions, as well as sequential and parallel tool calls with real tool execution.
Training used the Kimi-K2 configuration of the dataset. A total of 9,168 samples were used for a single epoch, taken from all four subsets in the following distribution:
| Subset | Samples | Share |
|---|---|---|
| single-turn-original | 2,750 | 30% |
| single-turn-diversify | 2,292 | 25% |
| irrelevant | 1,375 | 15% |
| multi-turn | 2,751 | 30% |
python
from transformers import AutoTokenizer, AutoModelForCausalLMimport torch model_id = "Rumiii/Qwimi-4B" tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.float16, device_map="cuda") messages = [ {"role": "user", "content": "What is the capital of France?"}] prompt = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True) inputs = tokenizer(prompt, return_tensors="pt").to(model.device) outputs = model.generate( **inputs, max_new_tokens=200, do_sample=False) print(tokenizer.decode(outputs[0], skip_special_tokens=True))Only a portion of the full dataset was used. Samples longer than 4,096 tokens were excluded (no sample was truncated), as were multi-turn samples that contain tool calls but no declared tools. The trajectories follow the standard Qwen3 tool-calling chat template.
This distillation was performed through an existing dataset. The Kimi K2 responses were not generated live for this model; the trajectories come from Agent-Ark/Toucan-1.5M, which was already available on Hugging Face.
from transformers import AutoTokenizer, AutoModelForCausalLMimport torch model_id = "Rumiii/Qwimi-4B" tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.float16, device_map="cuda") messages = [ {"role": "user", "content": "What is the capital of France?"}] prompt = tokenizer.apply_chat_template( messages, tokenize=False, add_generation_prompt=True) inputs = tokenizer(prompt, return_tensors="pt").to(model.device) outputs = model.generate( **inputs, max_new_tokens=200, do_sample=False) print(tokenizer.decode(outputs[0], skip_special_tokens=True))