What GPU do I need to run Qwen/Qwen3.8-27B-FP8?
A 27.8B language model for chat and instruction-following. 27.8B parameters, published in F8_E4M3. View on Hugging Face
Qwen3.8-27B-FP8 is published by Qwen on Hugging Face, with 5,528,743 downloads and 742 likes to date. It's a Qwen3_5ForConditionalGeneration model built for image-text-to-text, published natively in F8_E4M3.
What Qwen3.8-27B-FP8 is
Qwen3.8-27B-FP8 is a 27.8B-parameter language model published by Alibaba (Qwen) on Hugging Face. It is released under Apache 2.0.
License note: fully permissive, no gating and no usage restrictions. Facts in this section are sourced from Qwen3.8-27B-FP8's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Chat assistants
- Instruction following
- Synthetic data generation
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| FP8 (native) | 25.9 GB | 31.0 GB | RTX 4080 Super | 1 | $0.338/hr |
| INT4 (quantized) | 12.9 GB | 15.5 GB | RTX A5000 | 1 | $0.176/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 Qwen3.8-27B-FP8 at its published (F8_E4M3) precision: 1× RTX 4080 Super, at $0.338/hr per GPU ($0.338/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
Qwen3.8-27B-FP8: KV cache by context length
The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.
Hybrid: only the full-attention layers cache per token; the other layers keep a fixed-size state. Cached per token: 65,536 bytes at 16-bit.
| Context | KV cache, one sequence |
|---|---|
| 4K | 0.25 GB |
| 32K | 2.0 GB |
| 128K | 8.0 GB |
| 256K (model maximum) | 16.0 GB |
Computed from the layer, head and window counts in the model's own configuration (Qwen/Qwen3.8-27B-FP8), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.
Qwen3.8-27B-FP8: common questions
Does Qwen3.8-27B-FP8 fit on a 32 GB GPU?
Yes. At FP8 (native) it needs 31.0 GB of VRAM, so a 32 GB card holds it with 1.0 GB to spare. A 24 GB card is not enough for it at FP8 (native).
Is Qwen3.8-27B-FP8 already quantized?
Yes. It is published in FP8, one byte per parameter, so the 31.0 GB figure is already a quantized footprint rather than a full-precision one. Only INT4 goes below it, at 15.5 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.
What is the least VRAM Qwen3.8-27B-FP8 can run in?
15.5 GB, at INT4 (quantized), which fits a 16 GB card, against 31.0 GB at FP8 (native). That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Does quantizing Qwen3.8-27B-FP8 lower the GPU bill?
Yes. At FP8 (native) the cheapest live fit is one RTX 4080 Super at $0.338/hr. At INT4 (quantized) it drops to one RTX A5000 at $0.176/hr, provided a quantized checkpoint exists for it.
How to run Qwen3.8-27B-FP8
Run Qwen3.8-27B-FP8 with vLLM
Generic example, not from the model's own docs: adjust flags (quantization, context length, parallelism) for your setup.
vllm serve Qwen/Qwen3.8-27B-FP8 --tensor-parallel-size 1Run Qwen3.8-27B-FP8 with Ollama
Verified against Ollama's own library listing.
ollama run qwen3.8:27bDeploy Qwen3.8-27B-FP8 on Aquanode
Aquanode has no one-click deploy template for Qwen3.8-27B-FP8; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (1× RTX 4080 Super or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU podWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Qwen3.8 models
- Qwen3.8-27B (27.8B, BF16)
- JEV-27B-VL (27.8B, BF16)
- Qwen3.8-27B-OBLITERATED (27.8B, BF16)
- Qwen3.8-27B-Uncensored-FP8 (27.8B, F8_E4M3)
- Qwen3.8-27B-Uncensored (27.8B, BF16)
Lineage
Based on Qwen3.8-27B.
Alternatives at this size
Other models for image-text-to-text within about a third of Qwen3.8-27B-FP8's 27.8B parameters, from other model lines.
- gemma-4-31B-it (31.3B, BF16)
- diffusiongemma-26B-A4B-it (25.8B, BF16)
- Muse-Glimmer-30B (29.8B, BF16)
- Infinity-Parser2-Pro (35.1B, BF16)
- InternVL2-26B (25.5B, BF16)
More on Qwen3.8-27B-FP8
Fits on a 32 GB GPU at FP8: every model that fits in 32 GB.
Fits on a 16 GB GPU at INT4: every model that fits in 16 GB.
Related reading: H100 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.
