grant/spark-control
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 5 Wiki Activity

v0.27.3:0 - cap image resolution (fix oversized-image 400); remove vision-check button

Browse Source 
A 12MP photo expands past vLLM's ~4096-image-token limit -> 400. Cap via
--mm-processor-kwargs max_pixels in the qwen36 recipe so big images auto-
downscale server-side for every /v1 consumer (verified live: 400->200).
Remove the v0.27.2 in-dashboard vision-check button per owner request; the
vision badge already signals capability.
This commit is contained in:
Keysat
2026-06-18 18:41:28 -05:00
parent 9a3bf9ed86
commit 1f359e3c79
6 changed files with 9 additions and 113 deletions
Download Patch File Download Diff File Expand all files Collapse all files
image/models.yaml
+5
View File
@@ -110,3 +110,8 @@ models:
- --load-format=fastsafetensors
- --enable-prefix-caching
- --kv-cache-dtype=fp8
# Cap image resolution: a large phone photo (e.g. 12MP) otherwise expands
# to ~11.8k vision tokens, blowing past vLLM's ~4096-image-token limit and
# getting rejected with a 400. ~2MP auto-downscales big images server-side
# (so every /v1 consumer is covered) while staying sharp enough for OCR.
- '--mm-processor-kwargs={"max_pixels": 2000000}'