Can Arc Pro B70 32GB run Llama-3.1-8B-Instruct?

Intel · 32 GB GDDR6 · 608 GB/s bandwidth · Meta · 8.0B · model ctx up to 128K

ONEAPI

Yes — the Arc Pro B70 32GB (32 GB GDDR6) can run Llama-3.1-8B-Instruct. At Q4 with the default 8K context it needs ≈ 7.0 GB VRAM (weights 4.4 GB + KV cache 1.6 GB + 1 GB runtime overhead), which fits within 32 GB. Decode at Q4/8K: ≈ 48.4 tok/s (estimated, batch 1). FP16 (≈ 18.6 GB) also fits, at ≈ 13.3 tok/s (estimated). Estimates come from memory-bandwidth math; rows tagged measured override estimates. Relative ranking is more reliable than absolute tok/s. Methodology.

Fit & speed by quant and context

Quant	Context	VRAM needed	Fits?	tok/s (decode)
Q4	4K	6.2 GB	Yes	48.4 est.
Q4	8K default	7.0 GB	Yes	48.4 est.
Q4	32K	11.8 GB	Yes	48.4 est.
FP16	4K	17.8 GB	Yes	13.3 est.
FP16	8K default	18.6 GB	Yes	13.3 est.
FP16	32K	23.4 GB	Yes	13.3 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	Arc Pro B70 32GB VRAM
Q4	4.4 GB	1.6 GB	1 GB	7.0 GB	32 GB
FP16	16.0 GB	1.6 GB	1 GB	18.6 GB	32 GB

Other GPUs that run Llama-3.1-8B-Instruct

• B200 192GB SXM — ≈636.4 tok/s Q4 est.
• MI300X 192GB — ≈421.6 tok/s Q4 est.
• H200 141GB SXM5 — ≈381.8 tok/s Q4 est.
• H100 80GB SXM5 — ≈266.5 tok/s Q4 est.
• Gaudi 3 128GB — ≈260.6 tok/s Q4 est.
• RTX 5090 32GB — ≈214.3 tok/s Q4 measured
• A100 80GB SXM4 — ≈162.2 tok/s Q4 est.
• RTX 4090 24GB — ≈145.0 tok/s Q4 measured

All GPUs for Llama-3.1-8B-Instruct →

Other models for the Arc Pro B70 32GB

• MiniCPM5-1B — ≈358.2 tok/s Q4 est.
• Qwen3.5-2B — ≈193.5 tok/s Q4 est.
• DeepSeek-R1-0528-Qwen3-8B-layer-mix-bpw-3.8-mlx — ≈168.2 tok/s Q4 est.
• Qwen3.6-35B-A3B — ≈129.0 tok/s Q4 est.
• Qwen3.5-35B-A3B — ≈129.0 tok/s Q4 est.
• GLM-4.7-Flash — ≈129.0 tok/s Q4 est.
• Qwen3-30B-A3B-Instruct-2507 — ≈129.0 tok/s Q4 est.
• Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 — ≈129.0 tok/s Q4 est.

All models on the Arc Pro B70 32GB →