{"repo":"NousResearch/hermes-agent","public_id":"10809","display_label":"Vision preprocessing token limits and latency fixes","summary":"These issues and PRs address overly verbose vision preprocessing output, hardcoded max_tokens limits in auxiliary vision calls, and related latency problems. Work spans making max_tokens configurable via config.yaml, keeping preprocessing prompts concise, and routing improvements for multimodal models.","subsystem":"agent","counts":{"members":10,"open_issues":0,"open_issues_distinct":0,"open_prs":0,"closed_issues":3,"merged_prs":2,"attempt_prs":5},"landed_fix":{"status":"merged_to_main","basis":"structural","pr":75253},"resolution":{"state":"partially_resolved","by":"pr-75253"},"freshness":{"state":"verified","last_verified_at":"2026-08-15T07:23:38+00:00","is_stale":false},"grouping":{"backbone":"STRUCTURAL_PARTIAL","link_density":0.17155555555555554},"fix_edges":[{"issue_number":10809,"pr_number":75253,"tier":"verified","basis":"closing_ref","confidence":0.95,"recoverable_from_timeline":true},{"issue_number":10809,"pr_number":81269,"tier":"verified","basis":"closing_ref","confidence":0.95,"recoverable_from_timeline":true},{"issue_number":80025,"pr_number":75253,"tier":"verified","basis":"superseded_ref","confidence":0.9,"recoverable_from_timeline":false}],"members":[{"kind":"issue","number":10809,"title":"Default vision pre-process prompt generates overly long descriptions (~2000 chars), significantly slowing down image-bearing requests on local models","url":null,"state":"closed"},{"kind":"issue","number":29590,"title":"[Bug/Enhancement] Hardcoded max_tokens and overly verbose prompt in vision_tools.py causes severe latency for reasoning models","url":null,"state":"closed"},{"kind":"issue","number":34087,"title":"Make max_tokens configurable for aux vision calls","url":null,"state":"closed"},{"kind":"pr","number":10852,"title":"fix(gateway): keep auto vision preprocess concise","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#81269","number":81269},{"kind":"pr","value":"#10809","number":10809},{"kind":"pr","value":"#75253","number":75253}],"category":"merged_via","implication":"superseded"},"merged":false},{"kind":"pr","number":28999,"title":"fix(gateway): retry auto-vision on transient failure (#28972)","url":null,"state":"closed","close_reason":{"refs":[{"kind":"commit","value":"fb403a3a730f2a4eab1cfc60ffe854f7ea7c4159","number":0},{"kind":"pr","value":"#56889","number":56889}],"category":"implemented_elsewhere","implication":"superseded"},"merged":false},{"kind":"pr","number":29600,"title":"fix(vision): honor auxiliary.vision.max_tokens from config in vision_tools","url":null,"state":"closed","close_reason":{"category":"implemented_elsewhere","implication":"superseded"},"merged":false,"is_fix_pr":true,"fix_tier":"suggested","fix_detail_withheld":true},{"kind":"pr","number":38004,"title":"fix: respect auxiliary.vision.max_tokens from config.yaml (#29590)","url":null,"state":"closed","close_reason":{"refs":[{"kind":"commit","value":"db472f5df","number":0}],"category":"implemented_elsewhere","implication":"superseded"},"merged":false,"is_fix_pr":true,"fix_tier":"suggested","fix_detail_withheld":true},{"kind":"pr","number":61187,"title":"Cap gateway vision preprocessing output","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#10852","number":10852},{"kind":"pr","value":"#10809","number":10809},{"kind":"pr","value":"#10852","number":10852}],"category":"duplicate_of","implication":"duplicate"},"merged":false},{"kind":"pr","number":75253,"title":"fix(vision): stop capping aux vision output with hardcoded max_tokens","url":null,"state":"merged","close_reason":{"category":"merged","implication":"superseded"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["tests/tools/test_browser_console.py","tests/tools/test_video_analyze.py","tests/tools/test_vision_tools.py","tools/browser_tool.py","tools/vision_tools.py"]},{"kind":"pr","number":81269,"title":"fix(gateway): keep auto vision preprocess concise (salvage #10852)","url":null,"state":"merged","close_reason":{"category":"merged","implication":"superseded"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["contributors/emails/[redacted:email]","gateway/run.py","tests/gateway/test_vision_preprocess.py"]}],"trust":{"default_tier":"verified","includes_suggested":false}}