← AI Hashrate 中文

Can Titan V 12GB run Qwen3.5-35B-A3B?

NVIDIA · 12 GB HBM2 · 652.8 GB/s bandwidth · Alibaba · 34.7B (active 3.0B) · MoE · model ctx up to 256K

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

No — Qwen3.5-35B-A3B at Q4 needs ≈ 20.12 GB even at 4K context, beyond the Titan V 12GB's 12 GB HBM2. It would only run with heavy CPU/disk offload. Decode at Q4/8K: ≈ 139.6 tok/s (estimated, batch 1). FP16 needs ≈ 70.48 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.

Cheapest GPU that can: H100 80GB SXM5 See fit details →or☁️ Rent on RunPod from ~$0.50/hr

Fit & speed by quant and context

QuantContextVRAM neededFits?tok/s (decode)
Q44K20.12 GBNo140.1 ± est.
Q48K default20.16 GBNo139.6 ± est.
Q432K20.39 GBNo136.5 ± est.
FP164K70.44 GBNo
FP168K default70.48 GBNo
FP1632K70.7 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 neededTitan V 12GB VRAM
Q419.09 GB0.08 GB1 GB20.16 GB12 GB
FP1669.4 GB0.08 GB1 GB70.48 GB12 GB

Other GPUs that run Qwen3.5-35B-A3B

All GPUs for Qwen3.5-35B-A3B →

Other models for the Titan V 12GB

All models on the Titan V 12GB →