Can the Tesla P40 24GB Run DeepSeek-R1-Distill-Llama 70B?
Offload
Not entirely. DeepSeek-R1-Distill-Llama 70B at Q4_K_M is a 42.5 GB download (Hugging Face file listing for bartowski/DeepSeek-R1-Distill-Llama-70B-GGUF), about 18.5 GB more than the Tesla P40 24GB's 24 GB, so the remainder has to be offloaded to system RAM. Generation speed is then set by system-memory and PCIe bandwidth rather than by the GPU, so no tokens-per-second figure is quoted for this pair. The smallest card in this matrix that holds it entirely is the RTX A6000 48GB (48 GB).
Buy the Tesla P40 24GB
No Amazon listing for this card is tracked today — see live eBay listings.
Published runs for DeepSeek-R1-Distill-Llama 70B on the Tesla P40 24GB
No published run for this pair has passed the checks yet.
More about this pair
How much VRAM does DeepSeek-R1-Distill-Llama 70B need?
The Q4_K_M file is 42.5 GB. Add roughly 1.5 GB for the runtime and an ~8K-token context, so about 44 GB of VRAM holds it comfortably. Longer contexts grow the KV cache and need more.
What speed does DeepSeek-R1-Distill-Llama 70B reach on the Tesla P40 24GB?
No published run of DeepSeek-R1-Distill-Llama 70B on the Tesla P40 24GB has been cited yet. Generation reads every weight once per token, so the Tesla P40 24GB's 346 GB/s of memory bandwidth divided by the 42.52 GB file puts the theoretical ceiling near 8 tokens per second once the whole model is resident. That is an upper bound, not a speed estimate: kernel, launch and compute overhead keep real runs below it.
Check another pair
DeepSeek-R1-Distill-Llama 70B 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 — Offload
- RTX 5060 Ti 16GB — Offload
- Intel Arc A770 16GB — Offload
- Radeon RX 9070 XT — Offload
- RTX 4070 Ti SUPER — Offload
- RTX 4080 — Offload
- RTX 5070 Ti — Offload
- RTX 5080 — Offload
- RTX 3090 — Offload
- Radeon RX 7900 XTX — Offload
- RTX 4090 — Offload
- RTX 5090 — Offload
- RTX A6000 48GB — Runs, 13.9 tok/s median
- RTX PRO 5000 Blackwell 48GB — Runs, 26.1 tok/s (1 run)
- Instinct MI210 64GB — Runs
- RTX PRO 6000 Blackwell 96GB — Runs, 28.4 tok/s (1 run)
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
- gpt-oss-20b — 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
How the checks work
The fit is the Q4_K_M 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.