GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Zhanli-Li
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Zhanli-Li
Model Tree
Qwen/Qwen3.6-35B-A3B
Input Modalities
Output Modalities
Supported Functionality
A standard transformers checkpoint — weights only, no optimizer state and no
training configuration. The layout is the base model's:
| architecture | Qwen3_5MoeForConditionalGeneration (model_type: qwen3_5_moe), bfloat16 |
| parameters | 35B total / 3B activated — 40 layers, 256 experts, top-8 routed |
| size on disk | 70.21 GB across 14 safetensors shards (31,666 tensors) |
| hidden size | 2048 (text) |
| vocabulary | 248,320 tokens |
| context | 262,144 tokens (max_position_embeddings) |
| attention | hybrid: 3 × linear_attention + 1 × full_attention per block |
| base model | Qwen/Qwen3.6-35B-A3B |
The architecture, config and weight layout are the base model's, so the official Qwen3.6 deployment recipe applies unchanged — only the model path changes.
bash
python -m sglang.launch_server \ --model-path Zhanli-Li/DataWorld-35B-A3B \ --port 8000 --tp-size 8 \ --mem-fraction-static 0.8 --context-length 262144 \ --reasoning-parser qwen3 --tool-call-parser qwen3_codersglang >= 0.5.10, vllm >= 0.19.0, or a transformers that knows
qwen3_5_moe. The qwen3 reasoning parser and qwen3_coder tool-call parser
are what the official card prescribes, and this checkpoint was trained on
tool-calling agent trajectories, so keep them enabled.model.visual.*
tensors ship in their own shard, so multimodal serving works as on the base
model; mtp.* is absent, so the base card's speculative-decoding recipes do
not apply here.SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 / VLLM_ALLOW_LONG_MAX_MODEL_LEN=1
with --context-length / --max-model-len raised to 1,010,000.python
from transformers import AutoModelForCausalLM, AutoTokenizer repo = "Zhanli-Li/DataWorld-35B-A3B" # or a local pathmodel = AutoModelForCausalLM.from_pretrained(repo, dtype="bfloat16", device_map="auto")tokenizer = AutoTokenizer.from_pretrained(repo)Both the base model and this fine-tune are released under the Apache License 2.0 —
see LICENSE and the
base model card.
python -m sglang.launch_server \ --model-path Zhanli-Li/DataWorld-35B-A3B \ --port 8000 --tp-size 8 \ --mem-fraction-static 0.8 --context-length 262144 \ --reasoning-parser qwen3 --tool-call-parser qwen3_coderfrom transformers import AutoModelForCausalLM, AutoTokenizer repo = "Zhanli-Li/DataWorld-35B-A3B" # or a local pathmodel = AutoModelForCausalLM.from_pretrained(repo, dtype="bfloat16", device_map="auto")tokenizer = AutoTokenizer.from_pretrained(repo)