NVIDIA Nemotron-Labs-3-Puzzle-75B-A9B-NVFP4

NVIDIAπŸ‡ΊπŸ‡Έ United States
active
Context window1000K tokens

Version History

Puzzle-75B-A9B-NVFP4major

Compressed variant of Nemotron-3-Super reduced from 120.7B/12.8B active to 75.3B/9.3B active parameters using Iterative Puzzle framework. Achieves 2Γ— throughput and 8Γ— H100 concurrency while maintaining benchmark accuracy.

Benchmark Scores

Full leaderboard β†’
BenchmarkScoreRankSource
GPQA78.0%118 of 165Hugging Face model card, Jul 2026 β†—
MMLU-Pro82.2%40 of 76Hugging Face model card, Jul 2026 β†—
LiveCodeBench79.9%22 of 48Hugging Face model card, Jul 2026 β†—

Rank is against every model in our database with a sourced score on that benchmark, each counted once with its best score.

Coverage

model releaseNVIDIA

NVIDIA releases Nemotron-Labs-3-Puzzle-75B, compressed from 120B to 75B parameters with 2Γ— throughput

NVIDIA has released Nemotron-Labs-3-Puzzle-75B-A9B, a compressed variant of Nemotron-3-Super that reduces the model from 120.7B total/12.8B active parameters to 75.3B total/9.3B active parameters. According to NVIDIA, the model achieves approximately 2Γ— higher server throughput on a single 8Γ—B200 node and increases sustainable 1M-token single-H100 concurrency from 1 request to 8 requests while maintaining strong accuracy across benchmarks.

2 min read