Can Tesla P4 8GB run Qwen3.5-2B?

NVIDIA · 8 GB GDDR5 · 192 GB/s bandwidth · Alibaba · 2.0B · model ctx up to 256K

CUDA

Yes — the Tesla P4 8GB (8 GB GDDR5) can run Qwen3.5-2B. At Q4 with the default 8K context it needs ≈ 2.15 GB VRAM (weights 1.1 GB + KV cache 0.05 GB + 1 GB runtime overhead), which fits within 8 GB. Decode at Q4/8K: ≈ 61.1 tok/s (estimated, batch 1). FP16 (≈ 5.05 GB) also fits, at ≈ 16.8 tok/s (estimated). 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	2.12 GB	Yes	61.1 ± est.
Q4	8K default	2.15 GB	Yes	61.1 ± est.
Q4	32K	2.3 GB	Yes	61.1 ± est.
FP16	4K	5.03 GB	Yes	16.8 ± est.
FP16	8K default	5.05 GB	Yes	16.8 ± est.
FP16	32K	5.2 GB	Yes	16.8 ± 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	1.1 GB	0.05 GB	1 GB	2.15 GB	8 GB
FP16	4.0 GB	0.05 GB	1 GB	5.05 GB	8 GB

Other GPUs that run Qwen3.5-2B

• B200 192GB SXM — ≈2545.5 tok/s Q4 est.
• MI300X 192GB — ≈1686.4 tok/s Q4 est.
• H200 141GB SXM5 — ≈1527.3 tok/s Q4 est.
• H100 80GB SXM5 — ≈1065.9 tok/s Q4 est.
• Gaudi 3 128GB — ≈1042.4 tok/s Q4 est.
• A100 80GB SXM4 — ≈648.8 tok/s Q4 est.
• RTX 5090 32GB — ≈570.2 tok/s Q4 est.
• RTX 4090 24GB — ≈320.7 tok/s Q4 est.

All GPUs for Qwen3.5-2B →

Other models for the Tesla P4 8GB

• MiniCPM5-1B — ≈113.1 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.
• Mistral-7B-Instruct-v0.3 — ≈17.0 tok/s Q4 est.

All models on the Tesla P4 8GB →