Can the Tesla P40 24GB Run gpt-oss-20b?
Runs
Yes. gpt-oss-20b at MXFP4 is a 12.1 GB download (Hugging Face file listing for ggml-org/gpt-oss-20b-GGUF), and the Tesla P40 24GB has 24 GB of VRAM, leaving about 11.9 GB for the KV cache and runtime. No published speed measurement for this pair passes SpecPicks' sanity checks yet, so no tokens-per-second figure is quoted.
Buy the Tesla P40 24GB
No Amazon listing for this card is tracked today — see live eBay listings.
Published runs for gpt-oss-20b on the Tesla P40 24GB
No published run for this pair has passed the checks yet.
More about this pair
How much VRAM does gpt-oss-20b need?
The MXFP4 file is 12.1 GB. Add roughly 1.5 GB for the runtime and an ~8K-token context, so about 13.6 GB of VRAM holds it comfortably. Longer contexts grow the KV cache and need more.
What speed does gpt-oss-20b reach on the Tesla P40 24GB?
No published run of gpt-oss-20b on the Tesla P40 24GB has been cited yet. gpt-oss-20b is a mixture-of-experts model that reads about 2.08 GB of active weights per token, so the Tesla P40 24GB's 346 GB/s of memory bandwidth puts the theoretical ceiling near 166 tokens per second. That is an upper bound, not a speed estimate: kernel, launch and compute overhead keep real runs below it.
Check another pair
gpt-oss-20b on other GPUs
- RTX 4060 — Offload
- RTX 3060 Ti — Offload
- RTX 3060 12GB — Offload
- Intel Arc B580 — Offload
- RTX 4070 SUPER — Offload
- RTX 4060 Ti 16GB — Runs, 63.2 tok/s (1 run)
- RTX 5060 Ti 16GB — Runs, 102 tok/s median
- Intel Arc A770 16GB — Runs
- Radeon RX 9070 XT — Runs, 91.9 tok/s (1 run)
- RTX 4070 Ti SUPER — Runs, 106 tok/s median
- RTX 4080 — Runs, 138 tok/s median
- RTX 5070 Ti — Runs
- RTX 5080 — Runs, 134 tok/s median
- RTX 3090 — Runs
- Radeon RX 7900 XTX — Runs
- RTX 4090 — Runs
- RTX 5090 — Runs, 298 tok/s (1 run)
- RTX A6000 48GB — Runs
- RTX PRO 5000 Blackwell 48GB — Runs, 194 tok/s median
- Instinct MI210 64GB — Runs
- RTX PRO 6000 Blackwell 96GB — Runs
Other models on the Tesla P40 24GB
- Llama 3.2 1B — Runs, 97.8 tok/s median
- Llama 3.2 3B — Runs
- Mistral 7B — Runs, 31.5 tok/s median
- Qwen2.5 7B — Runs
- Llama 3.1 8B — Runs, 50.6 tok/s median
- DeepSeek-R1-Distill-Llama 8B — Runs
- Qwen3 8B — Runs
- Gemma 2 9B — Runs
- Gemma 3 12B — Runs
- Qwen2.5 14B — Runs, 14.5 tok/s median
- DeepSeek-R1-Distill-Qwen 14B — Runs
- Qwen3 14B — Runs
- Phi-4 14B — Runs
- Gemma 3 27B — Runs
- Gemma 2 27B — Runs
- Qwen3 30B-A3B — Runs
- Qwen3 32B — Runs
- Qwen2.5 32B — Runs
- DeepSeek-R1-Distill-Qwen 32B — Runs
- Qwen3 32B (Q6_K) — Offload
- Qwen2.5 32B (Q6_K) — Offload
- DeepSeek-R1-Distill-Qwen 32B (Q6_K) — Offload
- Gemma 3 27B (Q8_0) — Offload
- Llama 3.1 70B — Offload
- Llama 3.3 70B — Offload
- DeepSeek-R1-Distill-Llama 70B — Offload
How the checks work
The fit is the MXFP4 file size against the card's VRAM, with 1.5 GB kept for the runtime and an ~8K context. A published run counts only if the pair fits, the run is Q4-class, its source does not call it an estimate, it is under the bandwidth ceiling (memory bandwidth ÷ GB read per token: the whole file for a dense model, only the active experts' weights for a mixture-of-experts model such as gpt-oss-20b or Qwen3 30B-A3B, which is why those can post figures above bandwidth ÷ file size), it is not slower than the same card on a model twice its size, a single run is not slower than the same card on a larger dense model, and it is not one figure pasted across many cards. The whole matrix is downloadable as a CSV under CC BY 4.0.
As an Amazon Associate, SpecPicks earns from qualifying purchases. As an eBay Partner Network member, SpecPicks may earn a commission on eBay links.
This page is editorial synthesis based on publicly available information. No independent first-party benchmarking is reported.