Nemotron 3 Labs Ultra Math SFT

Public AI IQ model profile with source-backed benchmark results and derived ranking context.

IQ
n/a
EQ
n/a
Effective Cost
n/a
IQ Rank
n/a

Release and Capabilities

Public release2026-09-03
Modalitiestext
Context window1,048,576 tokens

Open weights for commercial and non-commercial use under OpenMDW-1.1.

Publisher-reported capabilities

  • Mathematical reasoning and proof-error identification checkpoint derived from Nemotron 3 Ultra.
  • The SFT checkpoint is independently downloadable under OpenMDW-1.1.

Limitations and availability caveats

  • The reported IMO gold-level performance belongs to an ensemble system, not this checkpoint alone.
  • Weights use OpenMDW-1.1; review the publisher license terms.

Official release and model references

Measured Coverage and Estimates

Nemotron 3 Labs Ultra Math Sft has 0 of 39 tracked IQ benchmark inputs across 0 of 7 dimensions. Raw benchmark tables contain source-backed results only. Derived IQ combines all six dimensions using the shared scoring methodology, including estimates where direct coverage is incomplete.

Benchmark versions, evaluation harnesses, and coverage affect comparisons; a missing result is not a score of zero. See the scoring methodology for estimation and uncertainty limits.

IQ Dimensions

Benchmark Results

BenchmarkScoreDimension

Compare

Frequently Asked Questions

What is the IQ of Nemotron 3 Labs Ultra Math SFT?
Nemotron 3 Labs Ultra Math SFT does not yet have an estimated IQ. AI IQ publishes a composite score once a model has source-backed benchmark results in at least two of its seven capability dimensions.
Is Nemotron 3 Labs Ultra Math SFT open source?
Yes — Nemotron 3 Labs Ultra Math SFT is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
When was Nemotron 3 Labs Ultra Math SFT released?
Nemotron 3 Labs Ultra Math SFT was publicly released on September 3, 2026.
How large is the context window of Nemotron 3 Labs Ultra Math SFT?
Nemotron 3 Labs Ultra Math SFT supports a context window of 1,048,576 tokens — the amount of text it can consider in a single request.