Text Generation
PEFT
Safetensors
English
Russian
qwen
qwen3.5
lora
unsloth
trl
sft
code
reasoning
bilingual
obsidian
graphite
conversational
Instructions to use OpenReasonAI/Graphite1.0-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use OpenReasonAI/Graphite1.0-4B with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-4B-Base") model = PeftModel.from_pretrained(base_model, "OpenReasonAI/Graphite1.0-4B") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Unsloth Studio
How to use OpenReasonAI/Graphite1.0-4B with Unsloth Studio:
Install Unsloth Studio (macOS, Linux, WSL)
curl -fsSL https://unsloth.ai/install.sh | sh # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for OpenReasonAI/Graphite1.0-4B to start chatting
Install Unsloth Studio (Windows)
irm https://unsloth.ai/install.ps1 | iex # Run unsloth studio unsloth studio -H 0.0.0.0 -p 8888 # Then open http://localhost:8888 in your browser # Search for OpenReasonAI/Graphite1.0-4B to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for OpenReasonAI/Graphite1.0-4B to start chatting
Load model with FastModel
pip install unsloth from unsloth import FastModel model, tokenizer = FastModel.from_pretrained( model_name="OpenReasonAI/Graphite1.0-4B", max_seq_length=2048, )
File size: 631 Bytes
f2661f1 497a6fd f2661f1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 | {
"model_id": "Qwen/Qwen3.5-4B-Base",
"train_seed": 3407,
"packing": false,
"eval_packing": false,
"completion_only_loss": false,
"assistant_only_loss": false,
"max_seq_length": 2048,
"learning_rate": 0.0001,
"lr_scheduler_type": "cosine",
"smoke_test": false,
"smoke_test_max_steps": 30,
"hub_push_enabled": true,
"hub_model_id": "Starred09/obsidian-critic-qwen35-4b-base-lora",
"hub_strategy": "checkpoint",
"resume_from_checkpoint": "/kaggle/working/obsidian_critic_qwen35_t4x2_unsloth/hub_resume/last-checkpoint",
"train_examples_after_filter": 36081,
"validation_examples_after_filter": 363
} |