feat(vertex): add validated Gemini model catalog for /model picker - #68767
Closed
ceverson70 wants to merge 1 commit into
Closed
ceverson70 wants to merge 1 commit into
ceverson70 wants to merge 1 commit into
Conversation
Add a "vertex" key to _PROVIDER_MODELS with 6 Gemini 3.x models that were validated live against the Vertex AI OpenAI-compatible endpoint (aiplatform.googleapis.com, global region, project antse-tooling) on 2026-07-21. All entries returned HTTP 200; 8 other candidates (e.g. gemini-3.1-flash, gemini-3-pro-preview) returned 404 and were excluded. Validated models (google/ prefix required by Vertex endpoint): - google/gemini-3.5-flash (frontier Flash, agentic) - google/gemini-3.6-flash (newer incremental over 3.5) - google/gemini-3.5-flash-lite (lighter/cheaper 3.5 variant) - google/gemini-3.1-pro-preview (3.1 Pro preview) - google/gemini-3-flash-preview (3.0 Flash preview) - google/gemini-3.1-flash-lite (most cost-efficient 3.x model) Context lengths are covered by the existing "gemini" prefix entry (1_048_576) in agent/model_metadata.py DEFAULT_CONTEXT_LENGTHS — no additional entries needed. Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
alt-glitch
added
type/feature
teknium1
added a commit
that referenced
this pull request
Jul 28, 2026
…vage Main gained its own vertex curated list (df051c1) two days after PR #68767 was opened, so the cherry-pick produced a duplicate 'vertex' dict key (later key silently wins in Python dict literals). Merge the two into one list: union of both, existing entries preserved, contributor's live-validated additions (gemini-3.6-flash, 3.5-flash-lite, 3.1-flash-lite) folded in.
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.
Collaborator
|
Merged via salvage PR #73516 with authorship preserved (d7f6dad). Main had independently grown its own vertex curated list (df051c1) two days after you opened this, so the two lists were merged in a follow-up commit — all three of your live-validated additions (gemini-3.6-flash, 3.5-flash-lite, 3.1-flash-lite) are in. Thanks @ceverson70! |
randlee
pushed a commit
to randlee/hermes-agent
that referenced
this pull request
Aug 11, 2026
…ch#68767 salvage Main gained its own vertex curated list (18d02c8) two days after PR NousResearch#68767 was opened, so the cherry-pick produced a duplicate 'vertex' dict key (later key silently wins in Python dict literals). Merge the two into one list: union of both, existing entries preserved, contributor's live-validated additions (gemini-3.6-flash, 3.5-flash-lite, 3.1-flash-lite) folded in.
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.
prmartinow
pushed a commit
to prmartinow/hermes-agent
that referenced
this pull request
Aug 26, 2026
…ch#68767 salvage Main gained its own vertex curated list (fc5ba3a) two days after PR NousResearch#68767 was opened, so the cherry-pick produced a duplicate 'vertex' dict key (later key silently wins in Python dict literals). Merge the two into one list: union of both, existing entries preserved, contributor's live-validated additions (gemini-3.6-flash, 3.5-flash-lite, 3.1-flash-lite) folded in.
prmartinow
pushed a commit
to prmartinow/hermes-agent
that referenced
this pull request
Aug 26, 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.
melon-xf
added a commit
to melon-xf/hermes-agent
that referenced
this pull request
Sep 3, 2026
…ch#68767 salvage Main gained its own vertex curated list (df051c1) two days after PR NousResearch#68767 was opened, so the cherry-pick produced a duplicate 'vertex' dict key (later key silently wins in Python dict literals). Merge the two into one list: union of both, existing entries preserved, contributor's live-validated additions (gemini-3.6-flash, 3.5-flash-lite, 3.1-flash-lite) folded in.
melon-xf
added a commit
to melon-xf/hermes-agent
that referenced
this pull request
Sep 3, 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.
Summary
"vertex"key to_PROVIDER_MODELSinhermes_cli/models.pywith 6 Gemini 3.x models validated live on the Vertex AI OpenAI-compatible endpoint (aiplatform.googleapis.com, global region).antse-toolingGCP project and returned HTTP 200.gemini-3.1-flash,gemini-3-pro-preview) returned 404 and were not included."gemini": 1_048_576prefix entry inagent/model_metadata.pyDEFAULT_CONTEXT_LENGTHSalready covers all Gemini models.Validated (HTTP 200):
google/gemini-3.5-flashgoogle/gemini-3.6-flashgoogle/gemini-3.5-flash-litegoogle/gemini-3.1-pro-previewgoogle/gemini-3-flash-previewgoogle/gemini-3.1-flash-liteRejected (HTTP 404):
gemini-3.1-flash,gemini-3.1-pro,gemini-3-pro-preview,gemini-3.5-pro,gemini-3.1-flash-lite-preview,gemini-3.1-flash-preview,gemini-3-flash,gemini-3-proTest plan
from hermes_cli.models import _PROVIDER_MODELS; _PROVIDER_MODELS["vertex"]returns the 6-model listantse-toolingproject (2026-07-21)/model vertexin hermes shows the picker with the 6 entries🤖 Generated with Claude Code