What GPU do I need to run ccore/Qwen3.5-4B-3D-Coder?
4.5B parameters, published in F16. View on Hugging Face
Qwen3.5-4B-3D-Coder is published by ccore on Hugging Face, with 289,466 downloads and 938 likes to date. It's a Qwen3_5ForConditionalGeneration model built for text-generation, published natively in F16.
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) |
|---|---|---|---|---|---|
| FP16 | 8.5 GB | 10.1 GB | V100 | 1 | $0.110/hr |
| FP8 (quantized) | 4.2 GB | 5.1 GB | RTX 5060 Ti | 1 | $0.180/hr |
| INT4 (quantized) | 2.1 GB | 2.5 GB | RTX 3070 | 1 | $0.088/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 Qwen3.5-4B-3D-Coder at its published (F16) precision: 1× V100, at $0.110/hr per GPU ($0.110/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
Qwen3.5-4B-3D-Coder: 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: 32,768 bytes at 16-bit.
| Context | KV cache, one sequence |
|---|---|
| 4K | 0.13 GB |
| 32K | 1.0 GB |
| 128K | 4.0 GB |
| 256K (model maximum) | 8.0 GB |
Computed from the layer, head and window counts in the model's own configuration (ccore/Qwen3.5-4B-3D-Coder), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.
Qwen3.5-4B-3D-Coder: common questions
Does Qwen3.5-4B-3D-Coder fit on a 12 GB GPU?
Yes. At FP16 it needs 10.1 GB of VRAM, so a 12 GB card holds it with 1.9 GB to spare. An 8 GB card is not enough for it at FP16.
What is the least VRAM Qwen3.5-4B-3D-Coder can run in?
2.5 GB, at INT4 (quantized), which fits a 6 GB card, against 10.1 GB at FP16. 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 Qwen3.5-4B-3D-Coder lower the GPU bill?
Yes. At FP16 the cheapest live fit is one V100 at $0.110/hr. At INT4 (quantized) it drops to one RTX 3070 at $0.088/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Qwen3.5 models
- NuExtract3 (4.5B, BF16)
- Qwen3.5-4B-GLB-3D-Coder (4.5B, F32)
- DeepSeek-V4-Pro-Qwen3.5-4B-UNCENSORED-heretic (4.5B, BF16)
- CogEvol-4B (4.5B, BF16)
- Qwen3.5-4B (4.7B, BF16)
Lineage
Based on Qwen3.5-4B.
More on Qwen3.5-4B-3D-Coder
Fits on a 12 GB GPU at FP16: every model that fits in 12 GB.
Fits on an 8 GB GPU at FP8: every model that fits in 8 GB.
Related reading: V100 pricing and specs, How much VRAM you need for LLMs, What AI inference is, and What AI model fine-tuning is.
Your own GPU cloud accounts
One console for the GPU clouds you already rent from. Your keys stay on your machine.
$20 a month, flat