{"repo":"NousResearch/hermes-agent","public_id":"91789","display_label":"GLM-5.3 graded reasoning_effort support fixes","summary":"These issues and pull requests address a bug where GLM-5.3 was incorrectly handled by a version-check function meant for GLM-5.2, causing the provider to send only binary thinking signals instead of the graded reasoning_effort values (low/medium/high/max) that GLM-5.3 supports.","subsystem":"agent/transports","counts":{"members":4,"open_issues":0,"open_issues_distinct":0,"open_prs":0,"closed_issues":1,"merged_prs":1,"attempt_prs":2},"landed_fix":{"status":"merged_to_main","basis":"structural","pr":91844,"merged_at":"2026-08-21T21:58:37+00:00"},"resolution":{"state":"resolved","by":"pr-91844"},"freshness":{"state":"verified","last_verified_at":"2026-08-21T21:58:37+00:00","is_stale":false},"grouping":{"backbone":"STRUCTURAL_SPANNING","link_density":0.5},"fix_edges":[{"issue_number":91789,"pr_number":91844,"tier":"verified","basis":"closing_ref","confidence":0.95,"recoverable_from_timeline":true,"pr_merged_at":"2026-08-21T21:58:37+00:00"}],"members":[{"kind":"issue","number":91789,"title":"[Bug] zai: glm-5.3 accepts graded reasoning_effort (low/medium/high/max) but provider plugin only sends binary thinking — _is_glm_5_2() misses glm-5.3","url":null,"state":"closed"},{"kind":"pr","number":91840,"title":"fix(zai): send graded reasoning_effort for glm-5.3, not just binary thinking","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#91844","number":91844},{"kind":"pr","value":"#91789","number":91789}],"category":"superseded_by_pr","implication":"superseded"},"merged":false},{"kind":"pr","number":91844,"title":"fix: GLM-5.3 low/medium reasoning effort no longer clamps to high (#91789)","url":null,"state":"merged","close_reason":{"category":"merged"},"merged":true,"is_fix_pr":true,"fix_tier":"verified","files_changed":["agent/reasoning_effort.py","plugins/model-providers/zai/__init__.py","tests/plugins/model_providers/test_zai_profile.py"]},{"kind":"pr","number":91965,"title":"fix(opencode): emit graded reasoning_effort for glm-5.3 on opencode-go","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#91844","number":91844}],"category":"implemented_elsewhere","implication":"superseded"},"merged":false}],"trust":{"default_tier":"verified","includes_suggested":false}}