beamster/Qwen3.8-Flash-Next-Sushi-2bpw: what it needs to run
beamster/Qwen3.8-Flash-Next-Sushi-2bpw is a quantization of Qwen/Qwen3.8-Flash-Next. Its configuration matches Qwen3.8-Flash-Next's on every field that decides memory, so Qwen3.8-Flash-Next's requirements are this repository's.
Answered by
Geometrypublished data. Both configurations are published files, read at the revisions named below and compared field by field. Not a test that the model loads.Qwen3.8-Flash-Next (Qwen/Qwen3.8-Flash-Next) — every field that decides memory is equal; what differs is noted below and does not change a memory answer.
Before you use those figures
- Its context limit is 1,048,576 tokens against the base's 262,144; the base's figures apply up to 262,144.
- Its safetensors index counts 21,142,110,099 parameters against the base's 179,999,981,459 (88.3% apart). The compared geometry is the same, so the difference is in weights this comparison does not cover, and the base's file sizes are off by about that share for this repository.
Qwen3.8-Flash-Next: weight-file size by format
| Format | Size | Range | Basis |
|---|---|---|---|
| FP16 | 360.2 GB | 252.0 GB – 363.6 GB | size range |
| Q8_0 | 191.4 GB | 133.9 GB – 194.4 GB | size range |
| Q6_K | 147.8 GB | 103.4 GB – 150.6 GB | size range |
| Q5_K_M | 128.6 GB | 86.6 GB – 150.6 GB | size range |
| Q5_0 | 125.6 GB | 86.6 GB – 150.6 GB | size range |
| Q4_K_M | 110.5 GB | 70.9 GB – 150.6 GB | size range |
| Q4_0 | 104.7 GB | 70.9 GB – 150.6 GB | size range |
| Q3_K_M | 90.0 GB | 54.1 GB – 150.6 GB | size range |
| Q2_K | 71.3 GB | 41.3 GB – 150.6 GB | size range |
These are Qwen3.8-Flash-Next’s sizes. A model with the same geometry quantizes to the same size, to within any difference in parameter count noted above. Which devices hold each, and how fast they run it, is on Qwen3.8-Flash-Next’s page.
What was read, and where
- beamster/Qwen3.8-Flash-Next-Sushi-2bpw at revision cb6c85e4f013 is tagged by its publisher as a quantization of Qwen/Qwen3.8-Flash-Next.
- beamster/Qwen3.8-Flash-Next-Sushi-2bpw/config.json at cb6c85e4f013 was compared with Qwen/Qwen3.8-Flash-Next/config.json at de4b8e4d43b9, the revision the catalogue pins: layers and their pattern, widths, head counts, experts, latent and recurrent dimensions, vocabulary, tied embeddings and any vision or audio tower.
Differences that do not change a memory answer
- max_position_embeddings: 1048576 here, 262144 in the base (context limit).
- Licence, as tagged
- other
- Parameters in its safetensors index
- 21B
- Task, as tagged
- image-text-to-text
- Last changed on Hugging Face
- 2026-10-07
Publishing this model? A badge for its card
[](https://llmbottleneck.com/hf/beamster/Qwen3.8-Flash-Next-Sushi-2bpw)
It states the memory to run the model at Q4_K_M with 8,192 tokens of context — built on this repository’s own Q4_K_M file when it publishes one — and links back to this page, where the evidence is.
Resolve another repository
The same answer as JSON: GET /v1/resolve?repo=beamster/Qwen3.8-Flash-Next-Sushi-2bpw, with a free key from /keys.