# NousResearch/hermes-agent · api_mode config ignored or hardcoded incorrectly

## Open now

- 1 issue, 2 PRs open on GitHub.
- #33600: provider: custom always appends /chat/completions to base_url, ignoring   api_mode
- PR #33604: fix(custom): normalize responses api mode alias
- PR #33889: feat: add api_mode=responses for generic OpenAI Responses API (#33600)

These members address a bug where api_mode from config.yaml is not respected for custom providers, and related refactoring to tie api_mode to provider config and clear it properly on provider switches.

## TL;DR

- 8 member(s): 1 open issue(s), 2 closed issue(s), 2 open PR(s), 0 merged PR(s), 3 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evidence (`STRUCTURAL_PARTIAL`): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.
- Members outside that structural core may belong to a different underlying problem.
- Pairwise link density `0.20`: 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

- #33600: provider: custom always appends /chat/completions to base_url, ignoring   api_mode

## Closed / duplicate issues

- #65031: Custom provider api_mode from config.yaml is ignored — always uses chat_completions
- #68159: custom_providers + api_mode: anthropic_messages silently routes to /v1/chat/completions instead of /v1/messages — usage.input_tokens always 0

## Prior & competing PRs

- **#33765: fix(provider): custom provider respects api_mode and works in cron/bg tasks (#33600, #33333)**. closed without merge · fix tier `suggested`
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `#33604`, `#33637`, `#33347`, `#33339`
  - _suggested-tier fix detail withheld; re-request with `?tier=all` to include it._
- **#33604: fix(custom): normalize responses api mode alias**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `0e4c879a3`
  - files: `agent/agent_init.py`, `hermes_cli/runtime_provider.py`, `tests/hermes_cli/test_runtime_provider_resolution.py`, `tests/run_agent/test_run_agent_codex_responses.py`
- **#33637: fix: accept responses as alias for codex_responses in api_mode config**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#33604`
- **#33884: feat: Prometheus — CommandCode provider, api_mode=responses, goal migration, Honcho prune, profile routing**. closed without merge
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `#33886`, `#33889`, `#33888`, `#33890`, `#33891`, `#33892`, `#33893`, `#32909`, `#33620`, `#29955`, `#18749`, `#18427`, `#33558`, `#33604`, `#33637`, `#33765`
- **#33889: feat: add api_mode=responses for generic OpenAI Responses API (#33600)**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`
  - files: `agent/chat_completion_helpers.py`, `hermes_cli/runtime_provider.py`

## Fix status

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