Add a multitask architecture module and its model sources - #1256
Merged
Conversation
…pare quantization scales with a tolerance
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 57cdd42. Configure here.
LeoGrin
force-pushed
the
leo/tabpfn-3-5
branch
from
September 9, 2026 22:12
26b074b to
a27bc6f
Compare
…, instead of a gating error
LeoGrin
force-pushed
the
leo/tabpfn-3-5
branch
from
September 9, 2026 22:19
a27bc6f to
42109c5
Compare
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Codex <noreply@openai.com>
priorphil
previously approved these changes
Sep 10, 2026
priorphil
left a comment
Collaborator
There was a problem hiding this comment.
I fixed Codex's comment and restored the simpler version of the "base" name check since the global was not used anywhere else.
priorphil
enabled auto-merge
September 10, 2026 07:46
jmkuebler
self-requested a review
September 10, 2026 07:48
Co-authored-by: Codex <noreply@openai.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.

Generated by Claude Code
Adds a new multitask architecture module and registers its two model versions in the loader, each backed by one checkpoint that serves both estimators. Nothing changes for the default estimator: the default model version stays
v3, the README is untouched, and the new versions are opt-in throughModelVersionor a file name. The checkpoints live in a private Hugging Face repo for now, and CI reaches them through the updatedHF_TOKEN.Commits and what they do
ModelVersion.V3_5/V3_5_FAST, model sources, CPU sample limit,create_default_for_versionwiring (Benjamin Jaeger, fromben/inference-update).create_default_for_versiontakes aModelVersionin the regressor, as the classifier already does (Benjamin Jaeger).20260909exports and list the multiclass variant.rtol=1e-6instead of exact equality. On macOS arm64 the two forward passes differ by 2 ulp in the absmax; the quantized keys and values are still compared exactly.tabpfn_v3_5_prereleasearchitecture alias and its test are dropped: the released checkpoints nametabpfn_v3_5.How this was checked
V3_5/V3_5_FAST.Not in this PR: making the new version the default, the README, and the docs. Those follow in a separate PR.
🤖 Generated with Claude Code