feat(curated): refresh FirstRun chat picks (qwen3 family + gpt-oss-20b)#27
Merged
Merged
Conversation
Old picks (qwen3-4b, llama32-3b, phi3-mini) were sized for sub-8 GB VRAM hosts. The reference deployment is Strix Halo with a ~100 GB unified memory pool — leaving most of it idle. This adds the 2026-05 chat refresh at the top of CURATED_MODELS so the wizard surfaces them first: qwen3-coder-next 49 GB frontier coder, full Strix Halo pool qwen3.6-27b 20 GB general-purpose default, Q5_K_XL gpt-oss-20b 12 GB OpenAI open-weights qwen3.5-9b 6 GB lean default — fits alongside embed/voice qwen3.5-0.8b 0.6 GB smoke-test pick, sub-second cold start qwen3-4b + phi3-mini stay in the featured set (explicit user ask) for mid-tier Vulkan hosts and the MIT-licensed pick. llama32-3b stays in the catalogue but moves below them; the wizard renders in list order. No tests churn — the old ids remain resolvable, so registry rows that reference them keep working. Follow-up (separate PR / session) will restructure the wizard into chat → embed/rerank → voice steps with FLM-multiplex detection. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
thinmintdev
force-pushed
the
feat/curated-chat-refresh-2026-05-16
branch
from
f2947da to
2667664
Compare
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Why
Old picks (qwen3-4b/llama32-3b/phi3-mini) were sized for sub-8 GB hosts. The reference deployment is Strix Halo with a ~100 GB unified pool, so we should default to picks that exercise it.
Notes / follow-ups
🤖 Generated with Claude Code