What GPU do I need to run Zyphra/Zamba2-1.2B-instruct?
1.2B parameters, published in F32. View on Hugging Face
Set up Zamba2-1.2B-instructZamba2-1.2B-instruct is published by Zyphra on Hugging Face, with 213,007 downloads and 32 likes to date. It's a Zamba2ForCausalLM model built for text-generation, published natively in F32.
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.
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 Zamba2-1.2B-instruct at its published (F32) precision: 1× V100 on simplepod, at $0.060/hr per GPU ($0.060/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
Zamba2-1.2B-instruct: common questions
How much VRAM does Zamba2-1.2B-instruct need?
5.4 GB at FP32, 1.4 GB at FP8 (quantized), 0.7 GB at INT4 (quantized). All of those sit under 6 GB, the smallest capacity this page reasons about, so VRAM is not what limits where this model runs. The 4.5 GB of weights plus inference overhead is the whole requirement.
Can Zamba2-1.2B-instruct run in 16-bit instead of FP32?
Yes. Its published weights are FP32, 4.5 GB, or 5.4 GB once inference overhead is added. Loading the same weights in 16 bits halves that to 2.3 GB, or 2.7 GB with overhead. How much accuracy the cast costs is model-specific and is not measured here.
How many copies of Zamba2-1.2B-instruct fit on one V100?
2, by VRAM alone. That card carries 16.0 GB and one copy needs 5.4 GB at FP32, on a live rate of $0.060/hr for the whole card. Throughput is not modelled here, so 2 copies is not 2 times the requests served.
Does quantizing Zamba2-1.2B-instruct lower the GPU bill?
Yes. At FP32 the cheapest live fit is one V100 on simplepod at $0.060/hr. At INT4 (quantized) it drops to one A16 on vultr at $0.059/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More Zyphra models
- Zamba-7B-v1 (7.2B, BF16)
- Zamba2-2.7B (2.7B, BF16)
- Zamba2-7B-Instruct (7.4B, BF16)
- bert-base-uncased (110M, F32)
- Qwen3-0.6B (752M, BF16)
- gpt2 (137M, F32)
