LLMBOTTLENECK.COM
Hugging Face repository / answered by Qwen3.5-35B-A3B

apodex/Apodex-1.1-mini: what it needs to run

apodex/Apodex-1.1-mini is a fine-tune of Qwen/Qwen3.5-35B-A3B. Its configuration matches Qwen3.5-35B-A3B's on every field that decides memory, so Qwen3.5-35B-A3B'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.

Qwen3.5-35B-A3B (Qwen/Qwen3.5-35B-A3B) — every compared field is stated by both and equal.

Qwen3.5-35B-A3B: weight-file size by format

FormatSizeRangeBasis
FP1671.9 GB50.3 GB – 72.6 GBsize range
Q8_038.2 GB26.7 GB – 39.5 GBsize range
Q6_K29.5 GB20.6 GB – 31.0 GBsize range
Q5_K_M25.7 GB17.3 GB – 31.0 GBsize range
Q5_025.1 GB17.3 GB – 31.0 GBsize range
Q4_K_M22.1 GB14.2 GB – 31.0 GBsize range
Q4_020.9 GB14.2 GB – 31.0 GBsize range
Q3_K_M18.0 GB10.8 GB – 31.0 GBsize range
Q2_K14.2 GB8.26 GB – 31.0 GBsize range

These are Qwen3.5-35B-A3B’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 Qwen3.5-35B-A3B’s page.

What was read, and where

  1. apodex/Apodex-1.1-mini at revision 4e4f109a5e32 is tagged by its publisher as a fine-tune of Qwen/Qwen3.5-35B-A3B.
  2. apodex/Apodex-1.1-mini/config.json at 4e4f109a5e32 was compared with Qwen/Qwen3.5-35B-A3B/config.json at 59d61f3ce65a, 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.
Licence, as tagged
apache-2.0
Parameters in its safetensors index
36B
Task, as tagged
text-generation
Last changed on Hugging Face
2026-08-26

Publishing this model? A badge for its card

The memory badge for this repository

[![Memory to run this model, from llmbottleneck.com](https://llmbottleneck.com/badge/apodex/Apodex-1.1-mini)](https://llmbottleneck.com/hf/apodex/Apodex-1.1-mini)

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=apodex/Apodex-1.1-mini, with a free key from /keys.

llmbottleneck
catalogue 2026-10-03models 327devices 135