Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to date.
ZwZ2-27B
inclusionAI·Instruct·released 16 Oct 2026
llama-server -hf inclusionAI/ZwZ2-27B-GGUF:Q4_K_MZwZ2-27B is the Qwen3.6-27B-based model in Zooming-without-Zooming-v2 (ZwZ2). It learns to connect visual evidence across regions through comparison, aggregation, matching, and multi-step reasoning. Training uses full images; inference requires no zooming tools.
How much VRAM does ZwZ2-27B need?
At Q4_K_M ZwZ2-27B needs about 19 GB with an 8K context (19 GB at Q4, 56 GB at full precision). Pick your GPU to see what fits and how fast it runs.
Needs: ≈19 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.
Advanced: context 8K · KV cache 16-bit
Smallest GPU for each quant and context (KV cache 16-bit)
| Quantisation | 8K | 16K | 32K | 64K | 128K | 256K |
|---|---|---|---|---|---|---|
| Q3_K_M | ||||||
| Q4_K_M | ||||||
| Q5_K_M | ||||||
| Q6_K | ||||||
| Q8_0 | ||||||
| FP16 |
≤16 GB card24–32 GB48–96 GBmore
| Quantisation | Total |
|---|---|
| Q2_KHeavy lossSmallest, noticeably worse quality | ≈11 GB |
| Q3_K_MNoticeable lossSmall, some quality loss | ≈16 GB |
| Q4_K_MrecommendedSmall lossRecommended default for local use | ≈19 GB |
| Q5_K_MNear-losslessNear-lossless for most tasks | ≈21 GB |
| Q6_KNear-losslessPractically lossless | ≈24 GB |
| Q8_0LosslessLossless in practice, twice the size of Q4 | ≈30 GB |
| FP16 / BF16LosslessFull precision, as released | ≈55 GB |
Estimates for 8K tokens of context: weights + KV cache + runtime overhead. Speed assumes memory-bound decoding at your GPU’s bandwidth. Pick a GPU above and the fit column checks every row against it. Weights marked “file” use the real GGUF download size; “est.” rows are computed from the parameter count.
How to pick a quant
Take the biggest quant that fits with room left for context. Q4_K_M is the usual sweet spot. Below Q4 quality drops fast, above Q6 you mostly pay memory for nothing. Big models (70B+, large MoE) tolerate low quants better than small ones.
- LosslessF16, Q8_0
- Near-losslessQ6_K, Q5_K_M
- Small lossQ5_K_S, Q4_K_M, IQ4_NL, IQ4_XS, Q4_K_S
- Noticeable lossQ4_0, Q3_K_L, Q3_K_M, IQ3_M
- Heavy lossIQ3_XXS, Q3_K_S, IQ2_M, Q2_K
Typical quality loss, not measured per model. Tiers come from KL divergence against FP16 on Llama 3 8B: llama.cpp perplexity README: LLaMA 3 8B scoreboard (KLD and ΔPPL vs FP16, Wikitext-2); llama.cpp quantize tool: ΔPPL per quant type @ Llama-3-8B
How much context fits?
At 8K context, Q4_K_M needs ≈19 GB(15 GB weights + 2.0 GB KV cache + 1.3 GB overhead). Pick a GPU above to see how much context fits.
Weights + overheadKV cache
Can the Tesla P40 run ZwZ2-27B? →
Speed vs quality per quant
Smaller quants run faster and fit in less memory, but lose quality.
Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.
Real-world speed of ZwZ2-27B
Tokens per second measured by people running it at home. A GPU shows a median once 3 reports agree on the same setup. Reports far from our estimate wait for a review. Measure yours with llama-bench →
No medians yet. Be one of the first 3 to report a setup.
All reports
Run ZwZ2-27B locally
--quantQ4_K_Mdefault; pick your GPU to tailor itinclusionAI/ZwZ2-27B-GGUF
llama-server -hf inclusionAI/ZwZ2-27B-GGUF:Q4_K_MDownloads the GGUF on first run. Installed through llama.app? The same flags work with `llama serve`.
vllm serve inclusionAI/ZwZ2-27Bservices:
vllm:
image: vllm/vllm-openai:latest
command: inclusionAI/ZwZ2-27B
ports:
- "8000:8000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 8000.
sglang serve inclusionAI/ZwZ2-27B --host 0.0.0.0 --port 30000services:
sglang:
image: lmsysorg/sglang:latest
command: sglang serve inclusionAI/ZwZ2-27B --host 0.0.0.0 --port 30000
ports:
- "30000:30000"
ipc: host
volumes:
- ~/.cache/huggingface:/root/.cache/huggingface
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]API on port 30000.
from transformers import pipeline
pipe = pipeline("image-text-to-text", model="inclusionAI/ZwZ2-27B", dtype="auto", device_map="auto")
messages = [
{
"role": "user",
"content": [
{"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
{"type": "text", "text": "Describe this image."},
],
}
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])Quantised versions and downloads
| Repository | Format | Quant | Size |
|---|---|---|---|
| prithivMLmods/inclusionAI-ZwZ2-27B-GGUF | GGUF | Q3_K_L | 14 GB |
| inclusionAI/ZwZ2-27B-GGUF | GGUF | Q4_K_M | 17 GB |
| prithivMLmods/inclusionAI-ZwZ2-27B-GGUF | GGUF | Q4_K_M | 17 GB |
| prithivMLmods/inclusionAI-ZwZ2-27B-GGUF | GGUF | Q5_K_M | 19 GB |
| prithivMLmods/inclusionAI-ZwZ2-27B-GGUF | GGUF | Q6_K | 22 GB |
| inclusionAI/ZwZ2-27B-GGUF | GGUF | Q8_0 | 29 GB |
| inclusionAI/ZwZ2-27B-GGUF | GGUF | F16 | 54 GB |
| prithivMLmods/inclusionAI-ZwZ2-27B-GGUF | GGUF | BF16 | 54 GB |
| JRamirez-UAB/ZwZ2-27B-EXL3-4.0bpw-24GB | EXL2 | 4.0BPW | — |
About ZwZ2-27B
ZwZ2-27B is the Qwen3.6-27B-based model in Zooming-without-Zooming-v2 (ZwZ2). It learns to connect visual evidence across regions through comparison, aggregation, matching, and multi-step reasoning. Training uses full images; inference requires no zooming tools. On 25 benchmarks across six domains, it improves over its base model on every benchmark, increasing overall accuracy from 55.52 to 60.51 (+4.99 points). Gains extend from fine-grained perception to spatial reasoning, long-context understanding, multimodal agents, and general visual reasoning. EvidenceConnect builds verified multi-region questions from SA-1B images: • Group 3–7 related regions and generate questions using privileged zoomed views. • Check answers through cross-model agreement and drop-one region tests. • Convert verified questions to four-option tasks and train with GRPO, without an additional supervised fine-tuning stage. All three model sizes use approximately 20K training examples. The policy sees full images with colored region outlines; privileged crop layouts are used only for data construction. See ZwZ2-RL-VQA for the data schema and synthesis process. Editable vector PDF. The figure shows eight representative benchmarks; the complete results follow.
Full model card on Hugging Face →Other ZwZ2 models
Whole family →Alternatives to ZwZ2-27B
All alternatives →Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
Following the February release of the Qwen3.5 series, we're pleased to share the first open-weight variant of Qwen3.6. Built on direct feedback from the community, Qwen3.6 prioritizes stability and real-world utility, offering developers a more intuitive, responsive, and genuinely productive coding…
Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
DiffusionGemma is a generative model built by Google DeepMind. Based on the 26B A4B Mixture-of-Experts (MoE) Gemma 4 architecture, DiffusionGemma generates tokens using discrete diffusion. This open-weights model is multimodal, handling text, image, and video inputs to generate text output.
ZwZ2-27B FAQ
- How much VRAM does ZwZ2-27B need?
- About 19 GB at Q4_K_M with an 8K context. Q8_0 needs about 31 GB and full precision about 56 GB. Longer contexts need more memory for the KV cache.
- Can ZwZ2-27B run on the Tesla P40?
- Yes. It fits at Q5_K_M (about 21 GB), at an estimated 9.8 tokens/s. It fits comfortably on the Radeon RX 7900 XTX, Instinct MI50 32GB, RTX 5090.
- Can I use ZwZ2-27B commercially?
- Unclear. It is released under Unknown licence. No licence stated: assume all rights reserved. Check the licence on the model card before using it in a product.
- What is the context length of ZwZ2-27B?
- 262,144 tokens (256K). The full window needs far more memory than an 8K context.
- How do I run ZwZ2-27B with llama.cpp?
- Run llama-server -hf inclusionAI/ZwZ2-27B-GGUF:Q4_K_M. It downloads the GGUF and starts an OpenAI-compatible server.
Rate this model
Data from Hugging Face, the model card and public leaderboards. Memory and speed figures are estimates; real numbers depend on the runtime, context length and drivers. Always check the licence yourself before commercial use.
0 comments