Skip to main content

Benchmark profile

Bulgarian Mathematical Olympiad 2025 (BRUMO 2025)

A challenging mathematical olympiad competition featuring problems that test advanced mathematical reasoning and problem-solving skills at the olympiad level.

Data verified

How BenchLM shows BRUMO 2025 right now

BenchLM is tracking BRUMO 2025 in the local dataset, but exact-source verification records for these rows are still being attached. To avoid a blank benchmark page, BenchLM shows the current tracked rows below as a display-only reference table.

These tracked rows are useful for inspection and spot-checking, but until exact-source attachments are completed they should not be treated as fully verified public benchmark rows.

106 tracked modelsLocal tracked rowsAwaiting exact-source attachmentsDisplay only

Tracked score on BRUMO 2025 — July 20, 2026

BenchLM mirrors the published tracked score view for BRUMO 2025. GPT-5.4 leads the public snapshot at 97% , followed by GPT-5.2 Pro (97%) and GPT-5.1-Codex-Max (96%). BenchLM does not use these results to rank models overall.

106 modelsMathCurrentDisplay onlyUpdated July 20, 2026

Tracked score table (106 models)

Score
1
GPT-5.4OpenAI
97%
2
97%
4
96%
5
96%
6
96%
8
96%
10
GPT-5.1OpenAI
96%
11
GPT-5.2OpenAI
96%
12
96%
13
96%
14
96%
15
96%
16
96%
17
96%
20
94%
21
93%
22
93%
23
92%
25
o3-proOpenAI
89%
26
89%
27
87%
28
o3OpenAI
87%
29
GLM-5Z.AI
87%
30
87%
32
85%
33
Qwen2.5-1MAlibaba
84%
34
84%
35
83%
36
83%
37
83%
38
82%
39
82%
40
80%
41
Mercury 2Inception
80%
42
79%
43
79%
44
78%
45
Kimi K2.5Moonshot AI
76%
46
75%
47
75%
48
Aion-2.0Aion Labs
73%
49
72%
51
72%
52
Seed 1.6ByteDance
71%
53
Seed-2.0-LiteByteDance
70%
54
69%
56
67%
57
67%
58
67%
60
GPT-4oOpenAI
65%
61
65%
63
64%
64
64%
66
63%
67
63%
69
61%
70
Seed-2.0-MiniByteDance
61%
71
Claude 3 OpusAnthropic
60%
72
59%
73
57%
75
55%
76
53%
77
Moonshot v1Moonshot AI
52%
78
51%
79
50%
80
49%
83
46%
85
LFM2-24B-A2BLiquidAI
45%
86
44%
87
DeepSeek-R1DeepSeek
43%
89
Nova ProAmazon
40%
91
38%
93
36%
94
35%
95
34%
97
32%
99
30%
100
29%
101
29%
102
28%
103
27%
105
23%
106
22%

The published BRUMO 2025 snapshot places GPT-5.4 first at 97%. The third row is 1.0 points behind. The broader top-10 range is 1.0 points, so many of the published results sit in a relatively narrow band.

106 models have been evaluated on BRUMO 2025. The benchmark falls in the Math category. This category carries a 5% weight in BenchLM.ai's overall scoring system. BRUMO 2025 is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About BRUMO 2025

Year

2025

Tasks

Olympiad problems

Format

Mathematical olympiad

Difficulty

Mathematical olympiad level

BRUMO represents the Bulgarian tradition of mathematical excellence, featuring problems that require deep mathematical insight and creative problem-solving approaches.

BenchLM freshness & provenance

Version

BRUMO 2025 2025

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 BRUMO 2025 measure?

A challenging mathematical olympiad competition featuring problems that test advanced mathematical reasoning and problem-solving skills at the olympiad level.

Which model leads the published BRUMO 2025 snapshot?

GPT-5.4 currently leads the published BRUMO 2025 snapshot with 97% tracked score. BenchLM shows this benchmark for display only and does not use it in overall rankings.

How many models are evaluated on BRUMO 2025?

106 AI models are included in BenchLM's mirrored BRUMO 2025 snapshot, based on the public leaderboard captured on July 20, 2026.

Last updated: July 20, 2026 · mirrored from the public benchmark leaderboard

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.