GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
HuggingFaceTB
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
HuggingFaceTB
Model Tree
HuggingFaceTB/nanowhale-100m-base
Input Modalities
Output Modalities
Supported Functionality
This model implements key DeepSeek-V4 innovations at a miniature scale:
| Component | Details |
|---|---|
| Parameters | ~110M total (41M embeddings, 69M non-embedding) |
| Hidden size | 320 |
| Layers | 8 |
| Attention heads | 8 (1 KV head — MQA-style) |
| MLA | Multi-head Latent Attention with q_lora_rank=160 |
| MoE | 4 routed experts + 1 shared, top-2 routing |
| Hyper-Connections | hc_mult=4, Sinkhorn routing (replacing residual connections) |
| MTP | 1 next-token prediction layer |
| Vocab | 129,280 (DeepSeek-V4 tokenizer) |
| Context | 2,048 tokens |
| Metric | Pretrained | SFT |
|---|---|---|
| Eval loss | — | 2.607 |
| Perplexity (held-out) | 13.62 | 12.90 |
| Token accuracy | 33.8% | 48.5% |
python
import torchfrom safetensors.torch import load_filefrom transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizerfrom huggingface_hub import hf_hub_download # Load model (recommended: manual load for reliability)config = AutoConfig.from_pretrained("HuggingFaceTB/nanowhale-100m", trust_remote_code=True)model = AutoModelForCausalLM.from_config(config, trust_remote_code=True).float() # Download and load weightsweights_path = hf_hub_download("HuggingFaceTB/nanowhale-100m", "model.safetensors")state_dict = load_file(weights_path)model.load_state_dict(state_dict, strict=True)model = model.cuda().eval() tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/nanowhale-100m") # Chatmessages = [{"role": "user", "content": "What are 3 benefits of exercise?"}]prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)input_ids = tokenizer.encode(prompt, return_tensors="pt").cuda()output = model.generate(input_ids, max_new_tokens=200, temperature=0.7, top_p=0.9, pad_token_id=tokenizer.eos_token_id)print(tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True))trust_remote_code=True.Trained on 1× NVIDIA H100 80GB.
Apache-2.0
import torchfrom safetensors.torch import load_filefrom transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizerfrom huggingface_hub import hf_hub_download # Load model (recommended: manual load for reliability)config = AutoConfig.from_pretrained("HuggingFaceTB/nanowhale-100m", trust_remote_code=True)model = AutoModelForCausalLM.from_config(config, trust_remote_code=True).float() # Download and load weightsweights_path = hf_hub_download("HuggingFaceTB/nanowhale-100m", "model.safetensors")state_dict = load_file(weights_path)model.load_state_dict(state_dict, strict=True)model = model.cuda().eval() tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/nanowhale-100m") # Chatmessages = [{"role": "user", "content": "What are 3 benefits of exercise?"}]prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)input_ids = tokenizer.encode(prompt, return_tensors="pt").cuda()output = model.generate(input_ids, max_new_tokens=200, temperature=0.7, top_p=0.9, pad_token_id=tokenizer.eos_token_id)print(tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True))