Oriel — AI & technology worth noticing, every day.

PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis

Paper · https://arxiv.org/abs/2610.20538

Multimodal code intelligence connects the symbolic world of programs with the visual world their execution creates. Interest in this capability is burgeoning, as models are increasingly asked to produce charts, webpages, animations, and semi-structured documents. Recently, proprietary models have pushed far ahead, producing interactive artifacts even from a text description or a screenshot alone. Yet their data and training remain closed, and open progress is held back by the quantity and quality of available data and by the pipelines and methods for producing it. Quality is the harder of the two, since a sample must be correct as a program, well built, and above all visually right once rendered. No existing signal covers all three. Execution often returns just a pass or fail, code similarity does not map to the visual output, and task-specific verifiers do not generalize. To push open progress further, we start from the data side with PrismaForge, a hierarchical data pipeline centered on a synthesized rubric that decomposes these factors into checkable criteria. The same rubric serves as both filter and reward: it curates the supervised corpus, then drives reinforcement learning on visual-coding tasks. Leveraging PrismaForge for data synthesis and filtering, we build PrismaCoder-2M, the most diverse open dataset for multimodal code intelligence and the largest to date. On the model side, we train PrismaCoder-9B and PrismaCoder-35B on this data and release both. They lead the open state of the art by a wide margin and rival leading proprietary systems. Our analyses further show that the proposed rubric-based pipeline improves multimodal coding performance across model backbones. We release our code and data at https://qiushisun.github.io/PrismaCoder-Home/.

First seen by Oriel on .

AI summary

PrismaCoder introduces a rubric-based data pipeline for multimodal code intelligence and releases a 2M-example dataset plus 9B and 35B models that lead open benchmarks and rival proprietary systems.

  • PrismaForge synthesizes hierarchical rubrics that break correctness, software quality, and visual fidelity into checkable criteria.
  • The rubrics filter the supervised training corpus and provide rewards for reinforcement learning on visual-coding tasks.
  • PrismaCoder-2M is described as the most diverse and largest open dataset for multimodal code intelligence to date.
  • The authors report that rubric-based training improves multimodal coding across model backbones; they release the code, dataset, PrismaCoder-9B, and PrismaCoder-35B.

#multimodal code #reinforcement learning #synthetic data #visual coding

Generated by gpt-5.6-sol. Check the original for details.

Metric trends

Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained

Points

Score reported by the source.

  • Hugging Face Papers: 72, latest observation ; +24 over observed period since

Comments

Comment count reported by the source.

  • Hugging Face Papers: 1, latest observation ; 0 over observed period since

Aliases

SourceExternal IDURL
Hugging Face Papers2610.20538https://arxiv.org/abs/2610.20538 Read discussion