FINAL-Bench / Qwen3.5-35B-A3B-VKAE

huggingface.co
Total runs: 0
24-hour runs: 0
7-day runs: 0
30-day runs: 0
Model's Last Updated: July 23 2026
text-generation

Introduction of Qwen3.5-35B-A3B-VKAE

Model Details of Qwen3.5-35B-A3B-VKAE

Qwen3.5-35B-A3B — VKAE Accelerated

Ready-to-run, VKAE-accelerated serving of Qwen3.5-35B-A3B (35B-parameter Mixture-of-Experts, ~3B active). Ships as a self-contained container — model weights and an optimized serving runtime in a single image — so anyone can reproduce the numbers on their own GPU.

VKAE (VIDRAFT Kernel Acceleration Engine) is VIDRAFT's proprietary inference-serving optimization. The acceleration recipe is withheld; only the reproducible results are published here.

Measured performance

NVIDIA B200, single GPU, FP8, same-harness before/after.

Metric Baseline VKAE Gain
Single-stream throughput 25.7 tok/s 601 tok/s 23.4×
Peak aggregate (high concurrency) — ~10,516 tok/s —
Output quality reference preserved no degradation

Realistic varied-content single-stream throughput sits around ~455 tok/s . Accuracy is preserved end to end.

Quick start
docker pull vidraft/qwen35-vkae:601
docker run --gpus all -p 8000:8000 vidraft/qwen35-vkae:601

The container serves an OpenAI-compatible API on port 8000 — point any OpenAI client at http://localhost:8000/v1 . A Blackwell (B200) or Hopper (H100/H200) class GPU is recommended.

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen35-vkae","messages":[{"role":"user","content":"Hello!"}]}'
Links
Base model & license

Base weights: Qwen/Qwen3.5-35B-A3B (Apache-2.0), unmodified. This card documents VIDRAFT's accelerated serving of the model; the base model itself is unchanged. The VKAE acceleration method is proprietary and is not distributed in source form.

Runs of FINAL-Bench Qwen3.5-35B-A3B-VKAE on huggingface.co

0
Total runs
0
24-hour runs
0
3-day runs
0
7-day runs
0
30-day runs

More Information About Qwen3.5-35B-A3B-VKAE huggingface.co Model

More Qwen3.5-35B-A3B-VKAE license Visit here:

https://choosealicense.com/licenses/apache-2.0

Qwen3.5-35B-A3B-VKAE huggingface.co

Qwen3.5-35B-A3B-VKAE huggingface.co is an AI model on huggingface.co that provides Qwen3.5-35B-A3B-VKAE's model effect (), which can be used instantly with this FINAL-Bench Qwen3.5-35B-A3B-VKAE model. huggingface.co supports a free trial of the Qwen3.5-35B-A3B-VKAE model, and also provides paid use of the Qwen3.5-35B-A3B-VKAE. Support call Qwen3.5-35B-A3B-VKAE model through api, including Node.js, Python, http.

Qwen3.5-35B-A3B-VKAE huggingface.co Url

https://huggingface.co/FINAL-Bench/Qwen3.5-35B-A3B-VKAE

FINAL-Bench Qwen3.5-35B-A3B-VKAE online free

Qwen3.5-35B-A3B-VKAE huggingface.co is an online trial and call api platform, which integrates Qwen3.5-35B-A3B-VKAE's modeling effects, including api services, and provides a free online trial of Qwen3.5-35B-A3B-VKAE, you can try Qwen3.5-35B-A3B-VKAE online for free by clicking the link below.

FINAL-Bench Qwen3.5-35B-A3B-VKAE online free url in huggingface.co:

https://huggingface.co/FINAL-Bench/Qwen3.5-35B-A3B-VKAE

Qwen3.5-35B-A3B-VKAE install

Qwen3.5-35B-A3B-VKAE is an open source model from GitHub that offers a free installation service, and any user can find Qwen3.5-35B-A3B-VKAE on GitHub to install. At the same time, huggingface.co provides the effect of Qwen3.5-35B-A3B-VKAE install, users can directly use Qwen3.5-35B-A3B-VKAE installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

Qwen3.5-35B-A3B-VKAE install url in huggingface.co:

https://huggingface.co/FINAL-Bench/Qwen3.5-35B-A3B-VKAE

Url of Qwen3.5-35B-A3B-VKAE

Qwen3.5-35B-A3B-VKAE huggingface.co Url

Provider of Qwen3.5-35B-A3B-VKAE huggingface.co

FINAL-Bench
ORGANIZATIONS

Other API from FINAL-Bench