Qwen3.8 27B
27BAlibaba Qwen3.8
Hybrid attention: only 16 of its 64 layers keep a KV cache, which is why a 27B model holds a 262K context in about 16GB of cache rather than 64GB. Sizing it as a conventional stack overstates the cache fourfold — the estimates here count the 16 and match published measurements at 8K, 32K and 262K. Quality loss per level has not been published.
262K
Max Context
1
Quant Variants
GGUF Q4_K_M
Best Quality
—
Accuracy Retained
Quantization Variants
Per-quant VRAM, quality loss, and inference speed on RTX 4090
Measured = run on this site · Estimated = not run here: calculated, or a figure this site has not verified · Community = public reports
Similar models
Compare with Seed-OSS 36BSeed-OSS 36B Instruct
ByteDance Seed
Dense 36B with a native 512K context. Q4 weights are ~22GB, which lands on a 24GB card or a 2×16GB split — but the context is the real cost: 512K of KV cache is roughly 128GB on its own, so budget context first and weights second.
GLM-4.7-Flash
Zhipu GLM-4.7
Zhipu's lightweight member of GLM-4.7: 30B total, about 3B active per token, MIT licence. Its attention is MLA, which caches one compressed 576-value vector per token per layer instead of full keys and values — so at 32K context the cache is under 2 GB where a conventional 47-layer model would need several times that. llama.cpp runs it as a DeepSeek-2-style model. Sizes below use the calculator's generic Q4_K_M/Q8_0 rates: no GGUF file size could be checked, and quality loss per level has not been published.
Gemma 4 26B-A4B IT
Google Gemma 4
Google's Gemma 4 mixture-of-experts model: about 25B parameters in the text model, roughly 4B active per token, a 256K window and image input. Only 5 of its 30 layers keep a cache that grows with context, and those use fewer, wider heads, so the cache stays small — about 0.9 GB at 32K and 5.3 GB at the full 256K. At Q4 that is about 17 GB at 32K: comfortable on a 24 GB card, which can still load the whole window (tight). A 16 GB card is just too small. Sizes use the calculator's generic rates — no GGUF file size could be checked, and no per-level quality loss has been published.
Gemma 4 31B IT
Google Gemma 4
Google's dense Gemma 4 flagship: about 31B parameters, a 256K window and image input. Only 10 of its 60 layers cache the full context, and those use four wide KV heads, so 128K of context adds about 11 GB of cache where a conventional 60-layer design would need about 120 GB. At Q4 it is about 21 GB at 4K — on a 24 GB card that is right at the edge of comfortable, and tight from 32K. A 32 GB card holds 32K comfortably. Sizes use the calculator's generic rates — no GGUF file size could be checked, and no per-level quality loss has been published.
Running Qwen3.8 27B locally
At Q4_K_M and 4K of context, Qwen3.8 27B needs about 17.5 GB — 15.6 GB of weights, 0.25 GB of KV cache and a 1.6 GB activation buffer. The smallest card in this index that clears that comfortably is the Radeon RX 7900 XT at 20 GB, and 32 of the 78 cards here do. These are calculated figures, not measurements: the estimate stops counting a card as comfortable at 88% of its VRAM, which is roughly the room a desktop session needs.
What longer context costs
Going from 4K to 32K adds about 1.75 GB, taking the total to 19.4 GB. That is gentler than the parameter count suggests: only 16 of 64 layers keep a KV cache that grows with context, so the cache scales at a fraction of the usual rate. The model's native window is 256K; holding all of it at Q4_K_M would need about 35 GB.
Which build to download
This index only tracks Qwen3.8 27B in GGUF, across 1 level (Q4_K_M). No per-level perplexity sweep has been published for this model, so the quality column is empty rather than estimated — within one model, more bits per weight is the only ordering the data supports. No throughput has been measured for this model on this site's hardware. GGUF runs on llama.cpp and Ollama across NVIDIA, AMD and Apple silicon; AWQ and GPTQ target vLLM on CUDA and ROCm; EXL2 is ExLlamaV2 — CUDA only, and now archived, so local use rather than serving.
Common questions
- How much VRAM does Qwen3.8 27B need?
- About 17.5 GB at Q4_K_M with 4K of context and batch 1, rising to roughly 19.4 GB at 32K. That figure is weights plus KV cache plus a 10% activation buffer, calculated from the model's architecture rather than measured on a card.
- Will Qwen3.8 27B run on a 20GB GPU?
- Yes — at Q4_K_M and 4K context it needs about 17.5 GB, which leaves 2.5 GB spare on a Radeon RX 7900 XT. That is the smallest card in this index that clears it comfortably; 32 of 78 do.
- Which quantization of Qwen3.8 27B should I use?
- Q4_K_M is the only level in the index for this model, so there is nothing to choose between here.
Where this model fits
Sized at Q4_K_M with a 4K context window, smallest card first. Comfortable means the estimate uses at most 88% of the memory.
- Runs comfortably on
- Radeon RX 7900 XT20GB · +2.5RTX 309024GB · +6.5Radeon RX 7900 XTX24GB · +6.5RTX 409024GB · +6.5Tesla P40 24G24GB · +6.5
- Tight but possible
- Mac M4 24G24GBMac M4 Pro 24G24GB
- Ships in
- GGUF
- Try it yourself
- Size it in the calculatorGenerate the run command
How to actually run this
Deployment guides for this model and this class of hardware.
This page's figures change when the model or the runtime does.Last updated 2026-10-20 RSS → /feed.xml