GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
CodeSoft
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
CodeSoft
Model Tree
Input Modalities
Output Modalities
Supported Functionality
Sorbet Mini Experimental was trained on 150M tokens, in 12 minutes, with one RTX 5060 Ti.
What matters is that a full pretrain run in twelve minutes means every change on the Sorbet line can be tested quickly without burning hours on a larger model.
Since Sorbet Mini is so cheap to train, it's a no brainer to keep training it. TinyStories was used to target basic language coherence as a starting point. Eventually, the full release, Sorbet Mini, will release and hopefully perform closer to other similarly sized models.
| knob | value |
|---|---|
| tokens | 149,999,872 (~37× Chinchilla) |
| steps | 1144 @ 512 seqs/step (seq 256) |
| data | TinyStories |
| precision | bf16, 8-bit AdamW |
| optimizer | AdamW lr 3e-4 → 1e-5 cosine, wd 0.1 (no decay on emb/norm), grad clip 1.0 |
| hardware | RTX 5060 Ti 16GB |
Result: train loss 8.13 → 2.81, val perplexity 3595 → 19.47.
bash
# very close to f16 (recommended)llama-completion -m sorbet-mini-experimental-q8_0.gguf \-p "Once upon a time," -n 128 --temp 0.8 --top-p 0.95# reference full-precision buildllama-cli -m sorbet-mini-experimental-f16.gguf -p "Hello, " -n 32
python
from transformers import AutoTokenizer, AutoModelForCausalLMimport torchtok = AutoTokenizer.from_pretrained("CodeSoft/sorbet-mini-experimental")model = AutoModelForCausalLM.from_pretrained("CodeSoft/sorbet-mini-experimental", dtype=torch.bfloat16)# model.to("cuda") # if you have a GPUprompt = "Once upon a time,"ids = tok(prompt, return_tensors="pt").input_idsout = model.generate(ids, max_new_tokens=128, do_sample=True,temperature=0.8, top_p=0.95)print(tok.decode(out[0], skip_special_tokens=True))