Can Tesla P4 8GB run Ornith-1.0-9B?

NVIDIA · 8 GB GDDR5 · 192 GB/s bandwidth · DeepReinforce · 9.0B · model ctx up to 256K

CUDA

Yes — the Tesla P4 8GB (8 GB GDDR5) can run Ornith-1.0-9B. At Q4 with the default 8K context it needs ≈ 7.75 GB VRAM (weights 4.95 GB + KV cache 1.8 GB + 1 GB runtime overhead), which fits within 8 GB. Decode at Q4/8K: ≈ 13.6 tok/s (estimated, batch 1). FP16 needs ≈ 20.8 GB — does not fit on this card. Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology. Buy Tesla P4 8GB on Amazon →

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	6.85 GB	Yes	13.6 ± est.
Q4	8K default	7.75 GB	Yes	13.6 ± est.
Q4	32K	13.15 GB	No	14.3 ± est.
FP16	4K	19.9 GB	No	1.7 ± est.
FP16	8K default	20.8 GB	No	1.6 ± est.
FP16	32K	26.2 GB	No	1.0 ± est.

Fits = weights + KV(ctx) + 1 GB overhead ≤ 95% of VRAM. Measured anchors are context-agnostic; the fit verdict is recomputed per context. A missing tok/s means the model is far beyond this card (offload-only territory).

VRAM breakdown at 8K context

Quant	Weights	KV cache	Overhead	Total needed	Tesla P4 8GB VRAM
Q4	4.95 GB	1.8 GB	1 GB	7.75 GB	8 GB
FP16	18.0 GB	1.8 GB	1 GB	20.8 GB	8 GB

Other GPUs that run Ornith-1.0-9B

• B200 192GB SXM — ≈565.7 tok/s Q4 est.
• MI300X 192GB — ≈374.7 tok/s Q4 est.
• H200 141GB SXM5 — ≈339.4 tok/s Q4 est.
• H100 80GB SXM5 — ≈236.9 tok/s Q4 est.
• Gaudi 3 128GB — ≈231.6 tok/s Q4 est.
• A100 80GB SXM4 — ≈144.2 tok/s Q4 est.
• RTX 5090 32GB — ≈126.7 tok/s Q4 est.
• RTX 4090 24GB — ≈71.3 tok/s Q4 est.

All GPUs for Ornith-1.0-9B →

Other models for the Tesla P4 8GB

• MiniCPM5-1B — ≈113.1 tok/s Q4 est.
• Qwen3.5-2B — ≈61.1 tok/s Q4 est.
• DeepSeek-R1-0528-Qwen3-8B-layer-mix-bpw-3.8-mlx — ≈53.1 tok/s Q4 est.
• Llama-3.2-3B-Instruct — ≈38.2 tok/s Q4 est.
• Qwen3-4B-Instruct-2507 — ≈30.5 tok/s Q4 est.
• NVIDIA-Nemotron-3-Nano-4B-BF16 — ≈30.5 tok/s Q4 est.
• Qwen3-4B-Thinking-2507 — ≈30.5 tok/s Q4 est.
• gemma-4-E2B-it — ≈24.0 tok/s Q4 est.

All models on the Tesla P4 8GB →