# Best LLMs for Math — September 2026 Leaderboard

> As of September 2026, Kimi K2.6 leads BenchLM's math leaderboard with a weighted score of 71.3.

- **Last verified:** September 4, 2026
- Canonical page: https://benchlm.ai/math
- **Ranking coverage:** 7 category-ranked models from 411 tracked models
- **Category weight:** 5% of the overall BenchLM score

## Current ranking

| Rank | Model | Creator | Weighted score | Published category rows | Exact-source rows (all categories) |
|------|-------|---------|----------------|----------------|-------------------|
| 1 | [Kimi K2.6](/models/kimi-2-6) | Moonshot AI | 71.3 | 5 | 34 total |
| 2 | [Claude Opus 4.8](/models/claude-opus-4-8) | Anthropic | 65.4 | 3 | 45 total |
| 3 | [GLM-5.1](/models/glm-5-1) | Z.AI | 64.1 | 6 | 31 total |
| 4 | [Qwen3.6 Plus](/models/qwen3-6-plus) | Alibaba | 62.4 | 7 | 53 total |
| 5 | [Kimi K2.5](/models/kimi-k2-5) | Moonshot AI | 62.4 | 9 | 49 total |
| 6 | [Claude Opus 4.5](/models/claude-opus-4-5) | Anthropic | 58.3 | 7 | 50 total |
| 7 | [GLM-5](/models/glm-5) | Z.AI | 56.9 | 8 | 41 total |

## Decision-ready shortlist

- #1 [Kimi K2.6](/models/kimi-2-6) — 71.3 weighted score, Open Weight, 256K context.
- #2 [Claude Opus 4.8](/models/claude-opus-4-8) — 65.4 weighted score, Proprietary, 1M context.
- #3 [GLM-5.1](/models/glm-5-1) — 64.1 weighted score, Open Weight, 203K context.
- #4 [Qwen3.6 Plus](/models/qwen3-6-plus) — 62.4 weighted score, Proprietary, 1M context.
- #5 [Kimi K2.5](/models/kimi-k2-5) — 62.4 weighted score, Open Weight, 256K context.

## Benchmarks in this category

### [AIME 2023](/benchmarks/aime2023) (American Invitational Mathematics Examination 2023)

A 15-question, 3-hour examination where each answer is an integer from 000 to 999. Serves as the intermediate step between AMC 10/12 and the USA Mathematical Olympiad (USAMO).

- Ranking status: Display only
- Year: 2023
- Format: Integer answers 000-999
- Difficulty: High school olympiad level

### [AIME 2024](/benchmarks/aime2024) (American Invitational Mathematics Examination 2024)

The 2024 edition of AIME, maintaining the same format of 15 challenging mathematics problems with integer answers from 000 to 999.

- Ranking status: Display only
- Year: 2024
- Format: Integer answers 000-999
- Difficulty: High school olympiad level

### [AIME 2025](/benchmarks/aime2025) (American Invitational Mathematics Examination 2025)

The most recent AIME examination, featuring 15 challenging mathematics problems testing olympiad-level mathematical reasoning with integer answers from 000-999.

- Ranking status: Display only
- Year: 2025
- Format: Integer answers 000-999
- Difficulty: High school olympiad level

### [AIME25 (Arcee)](/benchmarks/aime2025arcee) (AIME25 first-party comparison snapshot)

A display-only AIME25 reference from Arcee AI's Trinity-Large-Thinking launch chart.

- Ranking status: Display only
- Year: 2026
- Format: Integer answers 000-999
- Difficulty: High school olympiad level

### [HMMT Feb 2023](/benchmarks/hmmt2023) (Harvard-MIT Mathematics Tournament February 2023)

A prestigious high school mathematics competition hosted jointly by Harvard and MIT, featuring challenging problems across various mathematical disciplines.

- Ranking status: Display only
- Year: 2023
- Format: Competition mathematics
- Difficulty: High school olympiad level

### [HMMT Feb 2024](/benchmarks/hmmt2024) (Harvard-MIT Mathematics Tournament February 2024)

The 2024 February edition of the Harvard-MIT Mathematics Tournament, continuing the tradition of challenging high school mathematics competition.

- Ranking status: Display only
- Year: 2024
- Format: Competition mathematics
- Difficulty: High school olympiad level

### [HMMT Feb 2025](/benchmarks/hmmt2025) (Harvard-MIT Mathematics Tournament February 2025)

The most recent February edition of the Harvard-MIT Mathematics Tournament, featuring the latest challenging problems in competitive mathematics.

- Ranking status: Display only
- Year: 2025
- Format: Competition mathematics
- Difficulty: High school olympiad level

### [BRUMO 2025](/benchmarks/brumo2025) (Bulgarian Mathematical Olympiad 2025)

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

- Ranking status: Display only
- Year: 2025
- Format: Mathematical olympiad
- Difficulty: Mathematical olympiad level

### [MATH-500](/benchmarks/math-500) (MATH-500 Problem Set)

A curated subset of 500 problems from the MATH dataset, covering algebra, counting and probability, geometry, intermediate algebra, number theory, prealgebra, and precalculus.

- Ranking status: Display only
- Year: 2021
- Format: Free-form mathematical answers
- Difficulty: High school to undergraduate