# NousResearch/hermes-agent · Vision image preprocessing and size error handling

## Open now

- 1 issue, 2 PRs open on GitHub.
- #76505: \[Bug\] Native image_input_mode sends images at full resolution with no preprocessing — Qwen3VLProcessor rejects, but text-mode fallback succeeds with same image
- PR #76525: fix(vision): preprocess images before native image_input_mode (#76505)
- PR #131839: fix(vision): reactive image_too_large classification + GIF MIME fix (#76505) \[risk 0.80\]

These issues and PRs address a bug where images are sent at full resolution without preprocessing in native image input mode, causing rejections from Qwen3VLProcessor. The work involves fixing image preprocessing order, classifying resulting 400 errors as image_too_large, and correcting GIF MIME handling.

## TL;DR

- 4 member(s): 1 open issue(s), 0 closed issue(s), 2 open PR(s), 0 merged PR(s), 1 PR(s) closed without merging.
- No fix has been merged to main; one or more PRs were attempted (see below).
- **Recommended:** review the prior/competing PR(s) below. A fix was attempted but not merged; your case may extend or revive that work.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `unresolved`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_SPANNING`): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.
- Pairwise link density `0.50`: the share of member pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. `1.00` means every pair is linked.

## Open related issues

- #76505: \[Bug\] Native image_input_mode sends images at full resolution with no preprocessing — Qwen3VLProcessor rejects, but text-mode fallback succeeds with same image

## Prior & competing PRs

- **#76525: fix(vision): preprocess images before native image_input_mode (#76505)**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - files: `agent/conversation_compression.py`, `agent/conversation_loop.py`, `agent/error_classifier.py`, `run_agent.py`, `tests/run_agent/test_image_shrink_recovery.py`, `tools/vision_tools.py`
- **#131797: fix(agent): classify vLLM Qwen3VLProcessor 400 as image_too_large \[risk 0.30\]**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#76525`
- **#131839: fix(vision): reactive image_too_large classification + GIF MIME fix (#76505) \[risk 0.80\]**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `6b8614368fb34659a2c0e1ad99c37133da5cc24b`
  - files: `agent/conversation_compression.py`, `agent/error_classifier.py`, `agent/turn_recovery.py`, `tests/agent/test_image_shrink_recovery.py`, `tools/vision_tools.py`

## Fix status

- No merged fix on record.
- Subsystem: `agent`