NousResearch/hermes-agent

Compression threshold cap overrides user model_thresholds config

Cluster 117915 · agent. Data as of 2026-10-27 11:58 UTC

1 open item. Fix merged in #126064 on 28 Sep 2026. Check it covers your case.

open issues

open PR

items grouped

Known?: Yes. 8 related items grouped here, 1 issue already closed.
Fix event: A fix was merged to main in PR #126064. Check it covers your case. No release data is tracked.
Tracker: The maintainers' tracker marks this resolved, deciding PR #126064. That is separate from the fix event.
Still open: 0 issues and 1 PR.

Next step: read #126064, then check it covers your case.

Read #126064 Pick up a Lab task

Multiple fixes and a bug report addressing a default 256K token compression cap that silently overrides user-configured model_thresholds ratios for large-context models. Several approaches attempt to make explicit user configuration take precedence over the default cap.

Open issues0

No open issues.

Open PRs1

PR#117929fix(compression): let model ratios outrank the default cap

Fixes on record

Merged to main. A stamp counts the issues a fix is linked to.

#126064 Compaction follows compression.threshold again: no default 256K token cap (#117915, #125235)

merged to main

files: cli-config.yaml.example, hermes_cli/config_defaults.py, hermes_cli/config_migrations.py, tests/agent/test_context_compressor.py, tests/gateway/test_agent_cache.py, tests/tui_gateway/test_compression_config_hot_reload.py, tui_gateway/session_compression.py, website/docs/developer-guide/context-compression-and-caching.md, website/docs/user-guide/configuration.md, website/docs/user-guide/features/delegation.md

5 PRs closed without merging

From the close comments: done elsewhere 3 · superseded by another PR 2

#117961 fix(compression): don't apply default threshold_tokens cap when model_thresholds are explicitly set (closes #117915) · done elsewhere → #126064, #115986
#119247 fix(compression): don't apply default threshold_tokens cap when model_thresholds are explicitly set · superseded by another PR → dfe91e6, 95f2051, #119247
#119394 fix(compression): do not apply default threshold_tokens cap when model_thresholds are set · done elsewhere → #126064, #115986
#119441 Default 256K compression cap yields to user-set ratios (global threshold + model_thresholds) — salvage of #117929 · done elsewhere → #126064, #115986
#124012 fix(compression): surface the effective trigger and what binds it — startup line + hermes doctor (#117915) · superseded by another PR → 465c369

Closed issues · 1

Closed#117915[Bug]: default compression.threshold_tokens (256_000) silently overrides compression.model_thresholds ratio on 1M-window models

How this cluster was grouped

Link evidence (structural links throughout): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.
Pairwise link density 0.25: the share of item 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.

Pick up a Lab task

Check whether a fix covers the open issues here, with your own agent. Bring back evidence for maintainers.

Open the Lab

At a glance

Subsystem: agent
Merged PRs: #126064
Closed issues: 1
Attempts: 5 PRs closed without merging
Fix link checked: 28 Sep 2026

For your agent

The same facts, typed.

cluster/117915.json
cluster/117915.md