SDKs › vLLM: which chips it runs on
vLLM: which chips it runs on
vLLM and the chips it runs on, with evidence and the date each fact was last checked.
Snapshot 2026-10-15 · 394 configs on AI boxes and 423 on GPU cards · 123 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
vLLM project (vllm-project/vllm) · server. 55 chip entries: runs 32, partial 23.
- AMD Instinct MI300X (192GB): yes · evidence (checked 2026-10-07)
- AMD Instinct MI355X: yes · evidence (checked 2026-10-07)
- AMD Radeon RX 7900 XTX: yes · evidence (checked 2026-10-07)
- AMD Radeon RX 9070 XT: yes · evidence (checked 2026-10-07)
- AMD Strix Halo (Ryzen AI Max+ 395): yes · evidence (checked 2026-10-07)
- AWS Inferentia2: yes · evidence (checked 2026-10-07)
- AWS Trainium2: yes · evidence (checked 2026-10-07)
- Google Cloud TPU v6e: yes · evidence (checked 2026-10-07)
- Google Cloud TPU7x (Ironwood): yes · evidence (checked 2026-10-07)
- Huawei Ascend 910C: yes · evidence (checked 2026-10-07)
- Intel Gaudi 3: yes · evidence (checked 2026-10-07)
- NVIDIA A100 80GB: yes · evidence (checked 2026-10-07)
- NVIDIA B200: yes · evidence (checked 2026-10-07)
- NVIDIA B300 (Blackwell Ultra): yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 3060 12GB: yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 3090: yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 4060 Ti (16GB): yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 4070 Ti SUPER: yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 4080: yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 4080 SUPER: yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 5060 Ti (16GB): yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 5070 Ti: yes · evidence (checked 2026-10-07)
- NVIDIA GeForce RTX 5080: yes · evidence (checked 2026-10-07)
- NVIDIA H100 SXM: yes · evidence (checked 2026-10-07)
- NVIDIA H200: yes · evidence (checked 2026-10-07)
- NVIDIA L40S: yes · evidence (checked 2026-10-07)
- NVIDIA RTX 4090 (24GB): yes · evidence (checked 2026-10-07)
- NVIDIA RTX 5090 (32GB): yes · evidence (checked 2026-10-07)
- NVIDIA RTX 6000 Ada (48GB): yes · evidence (checked 2026-10-07)
- NVIDIA RTX A6000: yes · evidence (checked 2026-10-07)
- NVIDIA RTX PRO 6000 Blackwell (96GB): yes · evidence (checked 2026-10-07)
- Rebellions ATOM: yes · evidence (checked 2026-10-07)
- AMD Gorgon Halo (Ryzen AI Max+ PRO 495): partial · evidence (checked 2026-10-07)
- AMD Instinct MI325X: partial · evidence (checked 2026-10-07)
- Apple M-series Max (MacBook Pro / Mac Studio Max): partial · evidence (checked 2026-10-07)
- Apple M1 Max: partial · evidence (checked 2026-10-07)
- Apple M1 Ultra: partial · evidence (checked 2026-10-07)
- Apple M2 Max: partial · evidence (checked 2026-10-07)
- Apple M2 Ultra: partial · evidence (checked 2026-10-07)
- Apple M3 Max (40-core GPU): partial · evidence (checked 2026-10-07)
- Apple M3 Ultra: partial · evidence (checked 2026-10-07)
- Apple M4 Max (40-core GPU): partial · evidence (checked 2026-10-07)
- Apple M4 Pro: partial · evidence (checked 2026-10-07)
- Apple Mac Studio (M-series Ultra): partial · evidence (checked 2026-10-07)
- Huawei Ascend 310P: partial · evidence (checked 2026-10-07)
- Intel Arc B580 (12GB): partial · evidence (checked 2026-10-07)
- Intel Arc Pro B60 (24GB): partial · evidence (checked 2026-10-07)
- NVIDIA DGX Spark (GB10): partial · evidence (checked 2026-10-07)
- NVIDIA Jetson AGX Orin 64GB: partial · evidence (checked 2026-10-07)
- NVIDIA Jetson AGX Thor: partial · evidence (checked 2026-10-07)
- NVIDIA Jetson Orin Nano Super (8GB): partial · evidence (checked 2026-10-07)
- NVIDIA Jetson Orin NX 16GB: partial · evidence (checked 2026-10-07)
- Tenstorrent Blackhole p150: partial · evidence (checked 2026-10-07)
- Tenstorrent Wormhole n150: partial · evidence (checked 2026-10-07)
- Tenstorrent Wormhole n300: partial · evidence (checked 2026-10-07)
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.