GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
VertexResearch
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
VertexResearch
Model Tree
VertexResearch/Vertex-0.6-100M-8192-ctx-Base
Input Modalities
Output Modalities
Supported Functionality
| Parameters | 96.75M (tied embeddings) |
| Architecture | Qwen3-based transformer |
| Context length | 8192 (RoPE theta 1M) |
| Chat format | ChatML (<|im_start|> / <|im_end|>) |
| Tool calling | <tool_call> JSON blocks, system-prompt function definitions |
| EOS | </s> (2) and <|im_end|> (6) |
python
from transformers import AutoTokenizer, AutoModelForCausalLMrepo = "VertexResearch/Vertex-0.6-100M-8192-Instruct-v2"tok = AutoTokenizer.from_pretrained(repo)model = AutoModelForCausalLM.from_pretrained(repo)msgs = [{"role": "user", "content": "Who are you?"}]text = tok.apply_chat_template(msgs, add_generation_prompt=True, tokenize=False)ids = tok(text, return_tensors="pt", add_special_tokens=False).input_idsout = model.generate(ids, max_new_tokens=120, eos_token_id=[2, 6])print(tok.decode(out[0, ids.shape[1]:], skip_special_tokens=True))
For tool calling, put function definitions in the system prompt; the model emits
<tool_call>{"name": ..., "arguments": ...}</tool_call> and consumes results in
<tool_response> blocks.
Base: an 8192-context Vertex 0.6 100M base given a 1B-token quality anneal (FineWeb-Edu + synthetic elementary word-problem math). SFT with TRL on ~201K conversations: smol-smoltalk (~130K, concise multi-turn), UltraChat 200k (~40K, longer multi-turn), everyday-conversations (basic-chat grounding, 2×), function/tool-calling data, QA/tutoring, and self-identity. 2 epochs, lr 3e-4 cosine, bf16, max length 2048. Final eval loss 1.404.
These models are not the most coherent yet and need more tuning: expect rambling, repetition, and inconsistent answers, especially over longer generations.
97M parameters: multi-turn chat is much improved and tool-call syntax works, but factual accuracy is low, reasoning is shallow, and it makes arithmetic errors. Not for production use. Knowledge cutoff ~April 2024.
Tool calling works mechanically (correct <tool_call> format, stops cleanly)
with a single available function, but multi-step tool use is unreliable: with
more than one function available it can pick the wrong tool, and it can
hallucinate details when summarizing a tool's response rather than reporting it
accurately. Don't trust it in an unsupervised agent loop.