Qwen3.5-35B-A3B TIC

Bit-exact lossless .tic compression of Qwen/Qwen3.5-35B-A3B.

Same weights. No quantization.

What is TIC?

Precision BF16
Type multimodal, MoE
Raw weights 71.90 GB
.tic weights 51.27 GB (28.7% savings)

Benchmarks

See Benchmarks.

Run

# Install
curl -fsSL https://isiro.ai/install.sh | sh

# Download .tic bundle
pip install -U huggingface_hub
hf download isiroai/Qwen3.5-35B-A3B-TIC --local-dir Qwen3.5-35B-A3B-TIC

# Serve (vLLM supported today)
isiro serve Qwen3.5-35B-A3B-TIC --target vllm

Install options · Serve docs · Compile your own

Verify

Bit-exact checks. Hash manifest: model.tic.manifest.json.

# Confirm .tic integrity via the hash manifest
isiro verify Qwen3.5-35B-A3B-TIC

# Optional: confirm .tic weights match the raw weights (requires compiler)
hf download Qwen/Qwen3.5-35B-A3B --local-dir Qwen3.5-35B-A3B
isiro verify Qwen3.5-35B-A3B-TIC -r Qwen3.5-35B-A3B

License

Weights: same license as Qwen/Qwen3.5-35B-A3B (Apache-2.0).

.tic, ISIRO Runtime, and Compiler: ISIRO EULA.

Downloads last month
11
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support