What can I run on T4?
16 GB VRAMTuringMixed Precision (FP16/FP32) 65 FP16 TFLOPS; INT8 Precision 130 INT8 TOPS70 wattsView matrix for this target →
Turing has no BF16 support, so BF16 checkpoints must run in FP16 or FP32. Available on Hugging Face Jobs (t4-small, t4-medium).
Policy by policy
| Policy | Params | Inference memory (sourced, with precision) | On T4 | Best measured |
|---|---|---|---|---|
| LingBot-VA World model | 5.3B | ≥18 GB | T4 inferred · exceeds memory | needs ≥18 GB |
| ACT Imitation | 80M | 211 MB measured fp32 | no data | |
| Diffusion Policy Imitation | — | 1.12 GB measured fp32 | no data | |
| EO-1 VLA | 3B | ≥6.5 GB | no data | |
| Evo-1 VLA | 770M | — | no data | |
| Fast-WAM World model | 6B | — | no data | |
| GR00T N1.7 VLA | 3B | ≥16 GB | no data | |
| HIL-SERL RL | — | — | no data | |
| MolmoAct2 VLA | — | 12.1 GB measured bf16 | no data | |
| Multitask DiT Imitation | 450M | — | no data | |
| OpenVLA VLA | 7.5B | 15 GB measured bf16 | no data | |
| OpenVLA-OFT VLA | — | ≥16 GB | no data | |
| SmolVLA VLA | 450M | 1.75 GB measured fp32 | no data | |
| TD-MPC RL | — | — | no data | |
| VLA-JEPA VLA | — | — | no data | |
| VQ-BeT Imitation | — | — | no data | |
| WALL-OSS VLA | — | — | no data | |
| X-VLA VLA | 900M | — | no data | |
| π0 VLA | 3.3B | 13.32 GB measured fp32 | no data | |
| π0-FAST VLA | — | ≥8 GB | no data | |
| π0.5 VLA | — | ≥8 GB | no data |
“Exceeds memory” compares a policy's upstream-stated minimum inference memory (never a measured peak, which is precision-specific) with this target's VRAM. It is arithmetic, not observation, and is labelled inferred. Quantised or split runtimes can change the answer; a measured run always wins.
All measured runs on T4
No measured runs on T4 yet. A run needs runtime, precision, camera count. Contribute one.
Sources
- T1 · upstream documentedvendor-docsverified 2026-10-24 (1d ago)
T4 spec table: 65 FP16 TFLOPS, 130 INT8 TOPS, 16 GB GDDR6, 70 W.
“Mixed Precision (FP16/FP32) 65 FP16 TFLOPS INT8 Precision 130 INT8 TOPS INT4 Precision 260 INT4 TOPS Interconnect Gen3 x16 PCIe Memory Capacity 16 GB GDDR6 Bandwidth 320+ GB/s Power 70 watts”
https://www.nvidia.com/en-us/data-center/tesla-t4/ - T1 · upstream documentedvendor-docsverified 2026-10-24 (1d ago)
T4 page: based on the Turing architecture.
“Based on the new NVIDIA Turing ™ architecture”
https://www.nvidia.com/en-us/data-center/tesla-t4/ - T1 · upstream documentedofficial-docsverified 2026-10-24 (1d ago)
Hugging Face Jobs pricing lists T4 small and medium flavors (16 GB).
“Nvidia T4 - small 4 vCPU 15 GB 16 GB 50 GB $0.40 Nvidia T4 - medium 8 vCPU 30 GB 16 GB 100 GB $0.60”
https://huggingface.co/docs/hub/jobs-pricing