Skip to main content

Benchmark profile

Apex

A high-difficulty mathematical reasoning benchmark reported in DeepSeek-V4 model evaluations.

Data verified

Benchmark score on Apex — July 23, 2026

BenchLM mirrors the published score view for Apex. Qwen3.7 Max leads the public snapshot at 44.5% , followed by DeepSeek V4 Pro (Max) (38.3%) and DeepSeek V4 Flash (Max) (33.0%). BenchLM does not use these results to rank models overall.

9 modelsMathCurrentDisplay onlyUpdated July 23, 2026

Benchmark score table (9 models)

Score
1
Qwen3.7 MaxAlibaba · Closed
44.5%
2
DeepSeek V4 Pro (Max)DeepSeek · Open weight
38.3%
3
DeepSeek V4 Flash (Max)DeepSeek · Open weight
33.0%
4
ZAYA1-8BZyphra · Open weight
32.2%
5
DeepSeek V4 Pro (High)DeepSeek · Open weight
27.4%
6
Qwen3.7 PlusAlibaba · Closed
22.7%
7
DeepSeek V4 Flash (High)DeepSeek · Open weight
19.1%
8
DeepSeek V4 FlashDeepSeek · Open weight
1.0%
9
DeepSeek V4 ProDeepSeek · Open weight
0.4%

The published Apex snapshot places Qwen3.7 Max first at 44.5%. The third row is 11.5 points behind. The broader top-10 range is 44.1 points, so the table still separates the published systems.

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

About Apex

Year

2026

Tasks

Advanced mathematical reasoning

Format

Pass@1 math benchmark

Difficulty

Frontier math reasoning

BenchLM stores Apex as a display-only provider-table reference for exact first-party model comparisons.

BenchLM freshness & provenance

Version

Apex 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay 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 Apex measure?

A high-difficulty mathematical reasoning benchmark reported in DeepSeek-V4 model evaluations.

Which model scores highest on Apex?

Qwen3.7 Max by Alibaba currently leads with a score of 44.5% on Apex.

How many models are evaluated on Apex?

9 AI models have been evaluated on Apex on BenchLM.

Last updated: July 23, 2026 · BenchLM version Apex 2026

Choose a model with this week’s evidence

Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.

One email each week. Unsubscribe anytime.