← AI Hashrate 中文

Can Arc Pro B70 32GB run DeepSeek-V4-Flash?

Intel · 32 GB GDDR6 · 608 GB/s bandwidth · DeepSeek · 158.1B (active 13B) · MoE · model ctx up to 1000K

FF · No FitComposite score: VRAM headroom × 0.5 + tok/s speed × 0.5. Q4 @ 8K context.
ONEAPI

No — DeepSeek-V4-Flash at Q4 needs ≈ 88.12 GB even at 4K context, beyond the Arc Pro B70 32GB's 32 GB GDDR6. It would only run with heavy CPU/disk offload. Decode at Q4/8K: ≈ 11.2 tok/s (estimated, batch 1). FP16 needs ≈ 317.52 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.

Fit & speed by quant and context

QuantContextVRAM neededFits?tok/s (decode)
Q44K88.12 GBNo11.2 est.
Q48K default88.28 GBNo11.2 est.
Q432K89.25 GBNo10.9 est.
FP164K317.36 GBNo
FP168K default317.52 GBNo
FP1632K318.5 GBNo

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

QuantWeightsKV cacheOverheadTotal neededArc Pro B70 32GB VRAM
Q486.95 GB0.33 GB1 GB88.28 GB32 GB
FP16316.2 GB0.33 GB1 GB317.52 GB32 GB

Other GPUs that run DeepSeek-V4-Flash

All GPUs for DeepSeek-V4-Flash →

Other models for the Arc Pro B70 32GB

All models on the Arc Pro B70 32GB →