Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

Start free brief
BenchLM

FrontierMath legacy aggregate (FrontierMath (legacy))

Legacy FrontierMath values retained for historical model pages. This field is not used in current rankings because it can mix prior benchmark versions and slices.

Data verified 24 confirmed releases in the last 30 daysStart free brief

The public FrontierMath (legacy) snapshot ranks GPT-5.6 Sol first at 89%, ahead of GPT-5.6 Terra (84.9%) and GPT-5.6 Luna (78.6%) among 7 tested models. We mirror the table as display-only evidence; it does not affect overall rankings.

Benchmark score on FrontierMath (legacy) — August 22, 2026

We mirror the published score view for FrontierMath (legacy). GPT-5.6 Sol leads the public snapshot at 89%, followed by GPT-5.6 Terra (84.9%) and GPT-5.6 Luna (78.6%). We do not use these results to rank models overall.

1Closed

GPT-5.6 Sol

OpenAI

89%
Overall 81.73Context 1.05M
2Closed

GPT-5.6 Terra

OpenAI

84.9%
Overall 72.56Context 1.05M
3Closed

GPT-5.6 Luna

OpenAI

78.6%
Overall 66.96Context 1.05M
7 modelsMathRefreshingDisplay onlyUpdated August 22, 2026

Benchmark score table (7 models)

Score
1
GPT-5.6 SolOpenAI · Closed
89%
2
GPT-5.6 TerraOpenAI · Closed
84.9%
3
GPT-5.6 LunaOpenAI · Closed
78.6%
4
GPT-5.5 ProOpenAI · Closed
52.4%
5
GPT-5.5OpenAI · Closed
51.7%
6
GPT-5.4 ProOpenAI · Closed
50%
7
Claude Opus 4.7 (Adaptive)Anthropic · Closed
43.8%

The published FrontierMath (legacy) snapshot places GPT-5.6 Sol first at 89%. The third row is 10.4 points behind. The broader top-10 range is 45.2 points, so the table still separates the published systems.

7 models have been evaluated on FrontierMath (legacy). The benchmark falls in the Math category. This category carries a 5% weight in BenchLM.ai's overall scoring system. FrontierMath (legacy) is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About FrontierMath (legacy)

Year

2024

Tasks

Historical aggregate

Format

Open-ended mathematical reasoning with tool access

Difficulty

Research-level mathematics

Use the versioned FrontierMath v2 fields for ranking. They preserve the corrected v2 Tiers 1-3 and Tier 4 private-set scores separately rather than combining them.

FrontierMath: A Benchmark for Evaluating Advanced Mathematical Reasoning in AI

BenchLM freshness & provenance

Version

FrontierMath (legacy) 2024

Refresh cadence

Annual

Staleness state

Refreshing

Question availability

Public benchmark set

RefreshingDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does FrontierMath (legacy) measure?

Legacy FrontierMath values retained for historical model pages. This field is not used in current rankings because it can mix prior benchmark versions and slices.

Which model scores highest on FrontierMath (legacy)?

GPT-5.6 Sol by OpenAI currently leads with a score of 89% on FrontierMath (legacy).

How many models are evaluated on FrontierMath (legacy)?

7 AI models have been evaluated on FrontierMath (legacy) on BenchLM.

Compare Top Models on FrontierMath (legacy)

GPT-5.6 Sol vs GPT-5.6 TerraGPT-5.6 Terra vs GPT-5.6 LunaGPT-5.6 Luna vs GPT-5.5 ProGPT-5.5 Pro vs GPT-5.5
Last updated: August 22, 2026 · BenchLM version FrontierMath (legacy) 2024

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.