GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
bychwa
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
bychwa
Model Tree
Input Modalities
Output Modalities
Supported Functionality
I wanted a real end-to-end run I could finish in a weekend: niche data, custom tokenizer, pretrain a small causal LM, then LoRA-tune it for Q&A. Keeping the model tiny was intentional — fit the whole loop on one RTX 3090 pod, focus on process, and ship working artifacts.
| Architecture | GPT2LMHeadModel |
| Parameters | ~6.85M |
| Layers / emb / heads | 6 / 256 / 8 |
| Context | 256 tokens |
| Vocab | 8000 (ByteLevel BPE, trained on the recipe corpus) |
| Special tokens | <|pad|>, <|unk|>, <|bos|>, <|eos|>, <|user|>, <|assistant|> |
Weights are ~26 MB (safetensors).
Source: idoyaaran/mise-recipes (streamed from the Hub).
Each example was formatted roughly as:
text
<|bos|>{title}Ingredients: {ingredient names}Steps: {joined steps}<|eos|>
The weekend run used ~10k recipes (MAX_SAMPLES=10000 in the training script).
| Spec | Value |
|---|---|
| GPU | 1× NVIDIA GeForce RTX 3090 (24 GB) |
| CUDA | 13.0 |
| Host | Linux (RunPod container) |
| Python | 3.12 |
| Stack | PyTorch 2.5.1+cu121, Transformers 5.14, Datasets, Tokenizers, Accelerate, W&B |
Approximate cost: a few dollars at ~$0.25–0.40/hr for a short pretrain + SFT session.
Causal language modeling (next-token prediction) with Hugging Face Trainer.
| Hyperparameter | Value |
|---|---|
| Learning rate | 3e-4 |
| Warmup steps | 100 |
| Batch size | 16 |
| Grad accumulation | 4 (effective batch 64) |
| Epochs | 1 |
| Max length | 256 |
| Precision | fp16 |
| Optimizer | AdamW |
| LR schedule | linear |
| Metric | Value |
|---|---|
| Steps | 157 |
| Train runtime | ~24 s (this config on 3090) |
train_loss | ≈ 5.88 |
| Last logged step loss | ≈ 4.19 |
These are training metrics, not a held-out perplexity suite. Loss trended down; the model learns recipe-ish continuations, not general knowledge.
kitchenbot-chatNot intended as a general assistant, medical/nutrition advice source, or production kitchen system.
python
from transformers import AutoModelForCausalLM, AutoTokenizerrepo = "bychwa/kitchenbot-base"tok = AutoTokenizer.from_pretrained(repo)model = AutoModelForCausalLM.from_pretrained(repo)prompt = "<|bos|>Simple Tomato Sauce\nIngredients: tomatoes, garlic, olive oil\nSteps:"inputs = tok(prompt, return_tensors="pt")out = model.generate(**inputs, max_new_tokens=80, do_sample=True, temperature=0.8)print(tok.decode(out[0], skip_special_tokens=False))
For Q&A chat, load this base plus the LoRA adapter — see bychwa/kitchenbot-chat.
Full scripts (uv setup, corpus → tokenizer → pretrain → SFT → CLI):
https://github.com/bychwa/kitchenbot
bash
export HF_USER=bychwaexport WANDB_PROJECT=kitchenbotpython scripts/03_pretrain_base.py
Apache-2.0 for the model code/weights packaging in this card’s training setup. Respect the license/terms of idoyaaran/mise-recipes for the underlying text.
| Seed | 42 |