What GPU do I need to run unsloth/Qwen3.8-Flash-Next-FP8?
180.0B parameters, published in F8_E4M3. View on Hugging Face
Set up Qwen3.8-Flash-Next-FP8Qwen3.8-Flash-Next-FP8 is published by unsloth on Hugging Face, with 2,665 downloads and 33 likes to date. It's a Qwen4ExpForConditionalGeneration model built for image-text-to-text, published natively in F8_E4M3.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for KV-cache, activations, and fragmentation. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| FP8 (native) | 167.6 GB | 201.2 GB | RTX 4090 | 5 | $2.20/hr |
| INT4 (quantized) | 83.8 GB | 100.6 GB | RTX 5060 Ti | 7 | $0.770/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run Qwen3.8-Flash-Next-FP8 at its published (F8_E4M3) precision: 5× RTX 4090, at $0.441/hr per GPU ($2.20/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
Qwen3.8-Flash-Next-FP8: common questions
Can Qwen3.8-Flash-Next-FP8 run on a single GPU?
No. At FP8 (native) it needs 201.2 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest capable card in the live feed is a 48.0 GB RTX 4090, and it takes 5 of them.
Is Qwen3.8-Flash-Next-FP8 already quantized?
Yes. It is published in FP8, one byte per parameter, so the 201.2 GB figure is already a quantized footprint rather than a full-precision one. Only INT4 goes below it, at 100.6 GB. A GPU without FP8 tensor cores cannot run it as published, which is why cards here are matched on precision support and not on VRAM alone.
How many GPUs do I need to run Qwen3.8-Flash-Next-FP8?
5 at FP8 (native). It needs 201.2 GB of VRAM and the cheapest capable live offer is a 48.0 GB RTX 4090, so 5 of them come to $2.20/hr in total.
Does quantizing Qwen3.8-Flash-Next-FP8 lower the GPU bill?
Yes. At FP8 (native) the cheapest live fit is 5 RTX 4090 cards at $2.20/hr. At INT4 (quantized) it drops to 7 RTX 5060 Ti cards at $0.770/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Qwen3.8 models
- Qwen3.8-Flash-Next (180.0B, BF16)
- Qwen3.8-Flash-Next-FP8 (180.0B, F8_E4M3)
- Qwen3.8-Flash-Next-Uncensored-FP8 (180.0B, F8_E4M3)
- CYBER-FROST-3.8-BF16 (180.0B, BF16)
- Qwen3.8-Flash-Next (180.0B, BF16)
Related reading: RTX 4090 pricing and specs, and The best GPUs for AI, ranked.
