Particle 1.6 is a compact (~100M) chat model trained from scratch. It uses the same architecture and pretrained base as
Particle 1.0.
This release applies a second supervised fine-tune on an internal instruction dataset. That pass did not improve the model as much as expected. Everyday chat still works; factual reliability and consistency remain below what we wanted for a general-purpose assistant.
Weights are released under MIT. Training data is not included.
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3repo = "prathamkode/particle-1.6"
4tokenizer = AutoTokenizer.from_pretrained(repo)
5model = AutoModelForCausalLM.from_pretrained(repo)
6
7messages = [{"role": "user", "content": "hello"}]
8prompt = tokenizer.apply_chat_template(
9 messages, tokenize=False, add_generation_prompt=True
10)
11inputs = tokenizer(prompt, return_tensors="pt")
12output = model.generate(**inputs, max_new_tokens=64, do_sample=False)
13print(tokenizer.decode(output[0], skip_special_tokens=False))
The model is trained from random initialization. It is not a fine-tune of Llama, SmolLM, or any other public checkpoint.
The SFT mix is not published. It did not meet the quality bar we set for this release. Particle 1.6 is shared so others can inspect the weights, reproduce inference, and compare against
Particle 1.0.
Research, evaluation, and small demos. Suitable for studying from-scratch training at ~100M scale.
Not intended as a production assistant, a source of facts, or a coding model.
If you use these weights, please cite Particle and the public pretraining corpus used for the 1.0 base.
