NVIDIA Qwen3.6-35B-A3B-NVFP4 Card - SciCode: leaderboard
SciCode scientific research coding subproblems checked by unit tests. From NVIDIA's Qwen3.6-35B-A3B-NVFP4 model card (checkpoint released May 28, 2026): NVIDIA's evaluation of Alibaba's BF16 Qwen3.6-35B-A3B and its ModelOpt NVFP4 quantization with vLLM on GB300; the card does not state sampling, thinking mode or repeats. Accuracy (%), higher is better.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around February 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.6 35B A3B | 40.8 | #461 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=nvidia-qwen3-6-35b-a3b-nvfp4-card-scicode · How It Works · Data refreshed daily, snapshot 2026-11-12.