selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

tencent logo

Qwen3.8-27B-DSpine-b16

tencent·Base model·released 7 Nov 2026

Apache 2.02B256K ctx ≈1.9 GB
Your GPU
mlx_lm.server --model netprojects/Qwen3.8-27B-DSpine-b16-MLX --port 8080
Setup ↓

This checkpoint is a five-layer draft model for Qwen3.8-27B, with block size 16 and an 8,192-token sliding window.

Notify me when

How much VRAM does Qwen3.8-27B-DSpine-b16 need?

At Q4_K_M Qwen3.8-27B-DSpine-b16 needs about 1.9 GB with an 8K context (1.9 GB at Q4, 4.6 GB at full precision). Pick your GPU to see what fits and how fast it runs.

--vram
any
--gpus
--quant

Needs: ≈1.9 GB · Q4_K_M · 8K context — Pick a GPU to check the fit.

Change quant, context or KV cache to compare
Advanced: context 8K · KV cache 16-bit
Context
KV cache

Smallest GPU for each quant and context (KV cache 16-bit)

Quantisation8K16K32K64K128K256K
Q3_K_M
Q4_K_M
Q5_K_M
Q6_K
Q8_0
FP16

≤16 GB card24–32 GB48–96 GBmore

QuantisationQualityWeightsKV cacheTotal
Q2_KHeavy lossSmallest, noticeably worse qualityHeavy loss0.6 GB est.0.2 GB≈1.3 GB
Q3_K_MNoticeable lossSmall, some quality lossNoticeable loss0.9 GB est.0.2 GB≈1.6 GB
Q4_K_MrecommendedSmall lossRecommended default for local useSmall loss1.2 GB file0.2 GB≈1.9 GB
Q5_K_MNear-losslessNear-lossless for most tasksNear-lossless1.3 GB est.0.2 GB≈2.1 GB
Q6_KNear-losslessPractically losslessNear-lossless1.6 GB est.0.2 GB≈2.3 GB
Q8_0LosslessLossless in practice, twice the size of Q4Lossless2.0 GB file0.2 GB≈2.8 GB
FP16 / BF16LosslessFull precision, as releasedLossless3.8 GB file0.2 GB≈4.6 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?

Model quant
KV cache

At 8K context, Q4_K_M needs ≈1.9 GB(1.2 GB weights + 0.2 GB KV cache + 0.6 GB overhead). Pick a GPU above to see how much context fits.

024681K2K4K8K16K32K64K128K256KGB

Weights + overheadKV cache

Can the RTX 4060 run Qwen3.8-27B-DSpine-b16? →

Speed vs quality per quant

Smaller quants run faster and fit in less memory, but lose quality.

Q2_KQ3_K_MQ4_K_MQ5_K_MQ6_KQ8_0F16

Fits in VRAMTight fitCPU offload✕ won't runQuality is typical KL divergence vs FP16, not measured for this model.

Real-world speed of Qwen3.8-27B-DSpine-b16

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 Qwen3.8-27B-DSpine-b16 locally

These commands load the original weights from Hugging Face.

Apple Silicon with mlx-lmdocs
bash
pip install mlx-lm
mlx_lm.generate --model netprojects/Qwen3.8-27B-DSpine-b16-MLX --prompt "Hello"

Use `mlx_lm.chat` with the same --model for an interactive chat.

OpenAI-compatible serverdocs
mlx_lm.server --model netprojects/Qwen3.8-27B-DSpine-b16-MLX --port 8080

API on http://localhost:8080/v1.

Install MLX

Quantised versions and downloads

RepositoryFormatQuantSizeDownloads
aj9o9/Qwen3.8-27B-DSpine-b16-GGUFGGUFQ4_K_M1.3 GB19.8K
aj9o9/Qwen3.8-27B-DSpine-b16-GGUFGGUFQ8_02.2 GB19.8K
aj9o9/Qwen3.8-27B-DSpine-b16-GGUFGGUFBF164.1 GB19.8K
netprojects/Qwen3.8-27B-DSpine-b16-MLXMLX——1.9K

About Qwen3.8-27B-DSpine-b16

This checkpoint is a five-layer draft model for Qwen3.8-27B, with block size 16 and an 8,192-token sliding window.

Full model card on Hugging Face →

Alternatives to Qwen3.8-27B-DSpine-b16

All alternatives →
openbmb logo

MiniCPM5-2B

openbmb·Chat

≈2.3 GB

We are releasing MiniCPM5-2B, the second model in the MiniCPM5 series, following MiniCPM5-1B. It is a dense 2B Transformer that scales up the same training recipe, built for on-device, local deployment, and resource-constrained scenarios, reaching 2B-class open-source SOTA.

2.5B128K ctx Apache 2.0
1MSept 2026gguf
Qwen logo

Qwen3-1.7B

Qwen·Chat

≈2.6 GB

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

2B40K ctx Apache 2.0
3MApr 2025ollama
LiquidAI logo

LFM3-1.7B

LiquidAI·Chat

≈1.9 GB

LFM3-1.7B is a text model for local agents and specialized applications, based on our new tri-hybrid architecture.

1.7B256K ctx lfm1.0
100.6KOct 2026ollama
LiquidAI logo

LFM3-1.7B-Base

LiquidAI·Chat

≈1.9 GB

LFM3-1.7B-Base is the pre-trained checkpoint behind LFM3-1.7B, a text model for local agents and specialized applications based on our new tri-hybrid architecture.

1.7B256K ctx lfm1.0
11.6KOct 2026ollama
Qwen logo

Qwen3-1.7B-Base

Qwen·Chat

≈2.4 GB

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

1.7B32K ctx Apache 2.0
617.6KApr 2025gguf
LiquidAI logo

LFM2.5-2.6B

LiquidAI·Chat

≈2.6 GB

LFM2.5-2.6B is part of LFM2.5, a family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with a 128K context window and agentic post-training.

2.7B128K ctx lfm1.0
41.6KJul 2026gguf

Qwen3.8-27B-DSpine-b16 FAQ

How much VRAM does Qwen3.8-27B-DSpine-b16 need?
About 1.9 GB at Q4_K_M with an 8K context. Q8_0 needs about 2.8 GB and full precision about 4.6 GB. Longer contexts need more memory for the KV cache.
Can Qwen3.8-27B-DSpine-b16 run on the RTX 4060?
Yes. It fits at FP16 / BF16 (about 4.6 GB), at an estimated 37 tokens/s. It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.
Can I use Qwen3.8-27B-DSpine-b16 commercially?
Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.
What is the context length of Qwen3.8-27B-DSpine-b16?
262,144 tokens (256K). The full window needs far more memory than an 8K context.
How do I run Qwen3.8-27B-DSpine-b16 locally?
With MLX: mlx_lm.server --model netprojects/Qwen3.8-27B-DSpine-b16-MLX --port 8080.

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.