ZAYA1 VL 8B — Hardware Requirements & GPU Compatibility
Vision
ZAYA1-VL-8B is Zyphra's vision-language model of roughly 8 billion total parameters, built on its ZAYA1-8B language model with a Qwen2.5-VL vision encoder. The card describes two changes: vision-specific LoRA parameters that activate only on image tokens, and bidirectional attention across image tokens. It reports roughly 0.7 billion active parameters, was trained only on open data, and the card claims strong results for its size, such as 92.5 on DocVQA and 87.5 on AI2D. At this size, it fits on a single consumer GPU once quantized. The context window is 32,768 tokens. It is released under the Apache 2.0 license, permitting unrestricted commercial and research use. Published in April 2026, it extends the ZAYA1 base text model to image understanding.
43.7K downloads
42 likesSpecifications
- Publisher
- Zyphra
- Parameters
- 9.7B
- Architecture
- Zaya1VLForConditionalGeneration
- Context Length
- 32,768 tokens
- Vocabulary Size
- 262,272
- Release Date
- 2026-04-20
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does ZAYA1 VL 8B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 4.6 GB | 7.1 GB | 4.13 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 5.2 GB | 7.7 GB | 4.74 GB | 3-bit medium quantization |
| Q4_K_M | 4.80 | 6.3 GB | 8.8 GB | 5.83 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 7.4 GB | 9.9 GB | 6.93 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 8.5 GB | 11.0 GB | 8.02 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 10.2 GB | 12.7 GB | 9.72 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 19.9 GB | 22.4 GB | 19.44 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run ZAYA1 VL 8B?
Q4_K_M · 6.3 GBZAYA1 VL 8B (Q4_K_M) requires 6.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 9+ GB is recommended. Using the full 33K context window can add up to 2.5 GB, bringing total usage to 8.8 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3070 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run ZAYA1 VL 8B?
Q4_K_M · 6.3 GB58 devices with unified memory can run ZAYA1 VL 8B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, MacBook Air 13" M3 (8 GB).
Runs great
— Plenty of headroomWhere to Download ZAYA1 VL 8B
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does ZAYA1 VL 8B need?
ZAYA1 VL 8B requires 6.3 GB of VRAM at Q4_K_M, or 19.9 GB at BF16. Full 33K context adds up to 2.5 GB (8.8 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 9.7B × 4.8 bits ÷ 8 = 5.8 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 3 GB (at full 33K context)
VRAM usage by quantization
Q4_K_M6.3 GBQ4_K_M + full context8.8 GB- What's the best quantization for ZAYA1 VL 8B?
For ZAYA1 VL 8B, Q4_K_M (6.3 GB) offers the best balance of quality and VRAM usage. Q5_K_M (7.4 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 4.6 GB.
VRAM requirement by quantization
Q2_K4.6 GBQ4_K_M ★6.3 GBQ5_K_M7.4 GBQ6_K8.5 GBQ8_010.2 GBBF1619.9 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run ZAYA1 VL 8B on a Mac?
ZAYA1 VL 8B requires at least 4.6 GB at Q2_K, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run ZAYA1 VL 8B locally?
Yes — ZAYA1 VL 8B can run locally on consumer hardware. At Q4_K_M quantization it needs 6.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is ZAYA1 VL 8B?
At Q4_K_M, ZAYA1 VL 8B can reach ~762 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~104 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 6.3 × 0.65 = ~825 tok/s
Estimated speed at Q4_K_M (6.3 GB)
~825 tok/s~104 tok/s~825 tok/s~762 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of ZAYA1 VL 8B?
At Q4_K_M, the download is about 5.83 GB. The full-precision BF16 version is 19.44 GB. The smallest option (Q2_K) is 4.13 GB.
- Which GPUs can run ZAYA1 VL 8B?
52 consumer GPUs can run ZAYA1 VL 8B at Q4_K_M (6.3 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 40 GPUs have plenty of headroom for comfortable inference.
- Which devices can run ZAYA1 VL 8B?
59 devices with unified memory can run ZAYA1 VL 8B at Q4_K_M (6.3 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.