# FrontierScience Research leaderboard

Knowledge benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/knowledge-frontier-science-research

Every model the catalog carries a published FrontierScience Research value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Knowledge |
| Measure | Research evaluation |
| Tasks | Scientific research problems |
| Difficulty | Frontier scientific research |
| Published by | [Muse Spark Eval Methodology](https://ai.meta.com/static-resource/muse-spark-eval-methodology) |

A research-focused FrontierScience evaluation variant for scientific investigation and problem solving.

## FrontierScience Research ranking

5 models with a published FrontierScience Research value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Research evaluation | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Apodex 1.1](https://modelchoice.dev/models/apodex-1-1) | Apodex | 63.3 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Apodex 1.1 Mini](https://modelchoice.dev/models/apodex-1-1-mini) | Apodex | 51.7 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Sage-A1.5](https://modelchoice.dev/models/sage-a1-5) | InternScience | 48.3 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-5.4 Pro](https://modelchoice.dev/models/gpt-5-4-pro) | OpenAI | 36.7 | Observed 2026-11-21 · source benchlm:models |
| 5 | [Agents-A1-4B](https://modelchoice.dev/models/agents-a1-4b) | InternScience | 33.3 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
