fix(gemini): bump native provider aux default to gemini-3.6-flash - #70416
Closed
ildunari wants to merge 1 commit into
Closed
fix(gemini): bump native provider aux default to gemini-3.6-flash#70416ildunari wants to merge 1 commit into
ildunari wants to merge 1 commit into
Conversation
The native Gemini provider profile's default_aux_model and the curated model picker catalog were still pinned to gemini-3.5-flash, a stale generation now superseded by gemini-3.6-flash (documented GA). Bump both so the auxiliary-task default and the picker stay in sync with the current model. Contract test asserts the durable lockstep invariant only (default_aux_model is a member of _PROVIDER_MODELS["gemini"]) rather than pinning either side to a frozen model-name string, so it doesn't need updating on the next model-generation bump.
alt-glitch
added
type/bug
teknium1
added a commit
that referenced
this pull request
Jul 28, 2026
…l snapshot entries Follow-up to the #60063 salvage: the curated gemini list now carries gemini-3.6-flash (aux default, #70416) and the vertex list carries gemini-3.5-flash-lite (#68767) — both need snapshot pricing so direct Gemini/Vertex sessions don't report cost=unknown. Rates verified against https://ai.google.dev/gemini-api/docs/pricing (2026-07-28): 3.6-flash $1.50/$7.50, cache read $0.15; 3.5-flash-lite $0.30/$2.50, cache read $0.03.
Contributor
randlee
pushed a commit
to randlee/hermes-agent
that referenced
this pull request
Aug 11, 2026
…l snapshot entries Follow-up to the NousResearch#60063 salvage: the curated gemini list now carries gemini-3.6-flash (aux default, NousResearch#70416) and the vertex list carries gemini-3.5-flash-lite (NousResearch#68767) — both need snapshot pricing so direct Gemini/Vertex sessions don't report cost=unknown. Rates verified against https://ai.google.dev/gemini-api/docs/pricing (2026-07-28): 3.6-flash $1.50/$7.50, cache read $0.15; 3.5-flash-lite $0.30/$2.50, cache read $0.03.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem: the native Gemini provider profile's
default_aux_modeland the curated model-picker catalog (hermes_cli/models.py,_PROVIDER_MODELS["gemini"]) were both still pinned togemini-3.5-flash, superseded bygemini-3.6-flash— the documented, stable model code for the current generation.Fix: bump both in lockstep so the auxiliary-task default and the picker's offered model stay consistent.
Testing done: contract test (
tests/plugins/model_providers/test_gemini_profile.py) asserts the durable lockstep invariant —default_aux_modelis always a member of_PROVIDER_MODELS["gemini"]— rather than pinning either side to a frozen model string, so it won't need updating on the next model-generation bump. Confirmedtests/agent/test_auxiliary_client.py's existing gemini reasoning-wire-shape tests are unaffected (they pass an explicit literal model string, not the default).Other
gemini-3.5-flashreferences elsewhere inhermes_cli/models.pyare intentionally left untouched as out of scope for this minimal fix.