{"repo":"NousResearch/hermes-agent","public_id":"25989","display_label":"LM Studio context length and model loading issues","summary":"These issues and fixes address LM Studio provider bugs around incorrect context length detection, bypassing JIT loading, and models remaining loaded indefinitely. Changes include overhauling context resolution logic and fixing how models are preloaded and unloaded.","subsystem":"agent","counts":{"members":4,"open_issues":1,"open_issues_distinct":1,"open_prs":0,"closed_issues":2,"merged_prs":0,"attempt_prs":1},"landed_fix":null,"resolution":{"state":"unresolved","by":null},"freshness":{"state":"never_verified","last_verified_at":null,"is_stale":false},"grouping":{"backbone":"STRUCTURAL_PARTIAL","link_density":0.2866666666666666},"fix_edges":[],"members":[{"kind":"issue","number":25989,"title":"[Bug]: `lmstudio` provider pre-loads models manually, bypassing JIT and trampling user context config","url":null,"state":"open"},{"kind":"issue","number":66572,"title":"[Bug]: Hermes Agent always uses exactly 64000 tokens when INITIALIZING LM Studio model, ignoring configured context length","url":null,"state":"closed"},{"kind":"issue","number":67015,"title":"LM Studio JIT not used, Hermes leaves LM Studio models loaded indefinitely preventing loading of other models for later scheduled tasks","url":null,"state":"closed"},{"kind":"pr","number":48884,"title":"fix(lmstudio): use detected context for runtime preload when no override is set","url":null,"state":"closed","close_reason":{"refs":[{"kind":"pr","value":"#52188","number":52188},{"kind":"pr","value":"#45037","number":45037}],"category":"superseded_by_pr","implication":"superseded"},"merged":false,"is_fix_pr":true,"fix_tier":"suggested","fix_detail_withheld":true}],"trust":{"default_tier":"verified","includes_suggested":false}}