incoai / text-generation updated 3 days ago

Qwen3.8-27B-DFlash2

This repository contains the DFlash 2 draft model for Qwen/Qwen3.8-27B. It is not a standalone language model: it runs inside a speculative decoding server and drafts tokens for the target model to verify. The checkpoint is also mirrored at z-lab/Qwen3.8-27B-DFlash2.

Params
1.9 B
Context
262,144
Downloads 30d
391 K
Likes
246
Commercial use: allowed apache-2.0 Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 22 days
417 K203 K
Aug 31Sep 7Sep 14Sep 21

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

FileQuantSizeEst. VRAMVerdict on RTX 4090 · 24 GB
model.safetensors bf16 3.8 GB 5.0 GB ✅ Runs comfortably
model.safetensors (bf16, full) bf16 + 256K ctx 3.8 GB 14.0 GB ✅ Runs comfortably
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Run it

copy-paste, exact tags checked against the Hub
~ · curl · api/v1
$ curl -s https://aimodelscomparison.com/api/v1/models/qwen3-8-27b-dflash2
{
  "hf_id": "incoai/Qwen3.8-27B-DFlash2",
  "params_b": 1.92,
  "context_length": 262144,
  "license": { "id": "apache-2.0", "commercial": "yes" },
  "downloads_30d": 391228,
  "vram_estimates": [
    { "quant": "bf16", "gb": 5.0 }
  ],
  "updated_at": "2026-09-18T01:00:16Z"
}
est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture
DFlash2DraftModel
Parameters
1.9 B
Tensor type
BF16
Context length
262,144
Vocabulary
248,320
Layers / heads
5 / 32
Licence
apache-2.0
First seen on the Hub
2026-08-18
Base model
Qwen3.8-27B
Training datasets
undisclosed
Added to our catalog
2026-08-31

Family

Base model and the most-downloaded derivatives in the catalog.

Compare with any text-generation model