HU
HUST
Multimodal model ·Jul 2026

TurboVLA

Multimodal Open Source
200M
Parameters

About TurboVLA

TurboVLA is an AI model developed by HUST, in the multimodal category, released in 2026, made available as an open-source model with 0.2B params. TurboVLA is a real-time vision-language-action model at 32 Hz on an RTX 4090 with L->A pathway with direct V+L->A design. 97.7% average success on LIBERO with only 0.2B parameters.

On this page you'll find TurboVLA's full specifications. Review provider pricing and benchmark scores below, or compare TurboVLA head-to-head with other multimodal models.

→ More models by HUST → All multimodal models → AI benchmarks leaderboard

Run TurboVLA locally — what to buy

Share this build ↗

Sized to what TurboVLA actually needs (~8k context at Q4), not to the biggest GPU: Good = cheapest that runs it, Better = best value, Best = most headroom. Speed is a hardware estimate; anything past ~120 tok/s is shown as instant.

Best
Intel Arc B580 12GB
Consumer GPU · ~$249
~120 tok/s
Q4 · fits full context · up to 8k ctx
Runs with: ExLlamaV2, KoboldCpp, LM Studio
See full estimate →

Software support

·ExLlamaV2 ·KoboldCpp ·LM Studio ·MLX ·Ollama ·SGLang ·TensorRT-LLM ·llama.cpp ·vLLM ?KTransformers ?text-generation-webui

✓ measured · · compatible · — not supported. Informational only — speed is hardware-based.

Consumer GPU
120+ tok/s
39 options →
Apple
120+ tok/s
23 options →
Datacenter GPU
120+ tok/s
13 options →

Community

Would you run TurboVLA again?
No votes yet — be the first.
Did it run? — community reports by hardware

No reports yet. Own this model on some hardware? Be the first to confirm it runs.

Discussion

No comments yet. Start the discussion.

More models like TurboVLA

AI
NeuroChess
· 72.3K
 AQ
Qwen3.8 Max
Alibaba Qwen · 2.4T
 SA
Intern-S2 Preview 397B
Shanghai AI Lab · 397B
 SA
Apertus 1.5 70B
Swiss AI · 70B
 NVIDIA
Cosmos 3 Edge
NVIDIA
 OP
MiniCPM-RobotManip
OpenBMB · 1.5B

Frequently asked questions

What is TurboVLA? ▶

TurboVLA is a real-time vision-language-action model at 32 Hz on an RTX 4090 with L->A pathway with direct V+L->A design. 97.7% average success on LIBERO with only 0.2B parameters. It is tracked on GenAIList with its specifications, benchmark scores and provider pricing.

Who created TurboVLA? ▶

TurboVLA was developed by HUST and released in 2026.

Is TurboVLA open source or proprietary? ▶

TurboVLA is released as open source — its weights are publicly available to download, run locally and fine-tune.

How much does TurboVLA cost? ▶

Pricing for TurboVLA depends on the provider. See the providers table on this page for the latest API rates.

How does TurboVLA perform on benchmarks? ▶

Benchmark scores for TurboVLA are listed on this page as they are published. You can compare it head-to-head with other models on the GenAIList compare tool.

No videos yet. Know a good one? Add it below.

Know a great video about TurboVLA?