Beyond AIME
Seed2.1 Pro 87% leads since Jun 2026 Fading Models answer 100 original competition-style math problems at or above the hardest AIME level. Each answer is a positive integer checked by exact match.
ByteDance rewrote each problem from existing contests to reduce contamination.
Made by ByteDance Seed Released Apr 2025 Competition problems with answers All tasks public Last in a launch post Jun 2026 , dropped after it saturated
Last model added Jun 2026
- Level
- Competition The strongest students, in olympiad-style contests
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Short answer A number, word or short string matched to a key.
14
Models scored in lab cards
13
Launch posts list it, from 4 labs
8
System cards report it, from 3 labs
Seed2.1 Pro leads with 87% among models released by Jun 2026
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Jun 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 Seed2.1 Pro 87%
- 2 Seed2.0 Pro 86.5%
- 3 Seed2.0 Lite (0428) 79%
- 4 Seed1.8 77%
- 5
LongCat-Flash-Thinking 69.5% - 6 Sarvam 105B 69.1%
- 7
Mistral Medium 3.5 66.9% - 8 seed-oss-36b-instruct 65%
- 9
STEP3-VL-10B 63.2% - 10 Sarvam 30B 58.3%
- 11 Seed1.6-Thinking 56.3%
- 12 Seed-Thinking-v1.5 48%
- 13 LongCat-Flash-Omni 47.4%
- 14 LongCat-Flash-Chat 43%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Seed2.1 Pro | 87% | Jun 2026 | Model card | |
| 2 | Seed2.0 Pro | 86.5% | Feb 2026 | Model card | |
| 3 | Seed2.0 Lite (0428) | 79% | Apr 2026 | Model card | |
| 4 | Seed1.8 | 77% | Dec 2025 | Model card | |
| 5 | LongCat-Flash-Thinking | 69.5% | Sep 2025 | Model card | |
| 6 | Sarvam 105B | 69.1% | Mar 2026 | Model card | |
| 7 | Mistral Medium 3.5 | 66.9% | May 2026 | Model card | |
| 8 | seed-oss-36b-instruct | 65% | Aug 2025 | Model card | |
| 9 | STEP3-VL-10B | 63.2% | Jan 2026 | Model card | |
| 10 | Sarvam 30B | 58.3% | Mar 2026 | Model card | |
| 11 | Seed1.6-Thinking | 56.3% | Jun 2025 | Model card | |
| 12 | Seed-Thinking-v1.5 | 48% | Apr 2025 | Model card | |
| 13 | LongCat-Flash-Omni | 47.4% | Oct 2025 | Model card | |
| 14 | LongCat-Flash-Chat | 43% | Aug 2025 | Model card |
7 models have held number one since its release in Apr 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
ByteDance Seed Meituan
The best score passed 85% in Feb 2026
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Reported by the lab in its model card
Best so far
Labs headline it less often now, last in Jun 2026
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 23 Jun 2026 | Seed2.1 Pro | 87 | BeyondAIME |
| 22 May 2026 | Mistral Medium 3.5 | 66.9 | Beyond AIME avg@16 |
| 28 Apr 2026 | Seed2.0 Lite (0428) | 79 | BeyondAIME |
| 6 Mar 2026 | Sarvam 30B | 58.3 | Beyond AIME |
| 6 Mar 2026 | Sarvam 105B | 69.1 | Beyond AIME |
| 14 Feb 2026 | Seed2.0 Pro | 86.5 | BeyondAIME |
| 18 Dec 2025 | Seed1.8 | 77 | BeyondAIME |
| 31 Oct 2025 | LongCat-Flash-Omni | 47.4 | BeyondAIME(avg@10) |
| 21 Sep 2025 | LongCat-Flash-Thinking | 69.5 | BeyondAIME(Mean@10) |
| 29 Aug 2025 | LongCat-Flash-Chat | 43 | BeyondAIME(avg@10) |
| 20 Aug 2025 | seed-oss-36b-instruct | 65 | BeyondAIME |
| 25 Jun 2025 | Seed1.6-Thinking | 56.3 | BeyondAIME |
| 10 Apr 2025 | Seed-Thinking-v1.5 | 48 | Beyond AIME |
8 system cards report it, from 3 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 23 Jun 2026 | Seed2.1 Pro model card | 87 | BeyondAIME (Table 11) | Capabilities |
| 21 Mar 2026 | Seed1.8 model card | 77 | BeyondAIME (Table 1) | Capabilities |
| 14 Feb 2026 | Seed2.0 Pro model card | 86.5 | BeyondAIME (Table 3) | Capabilities |
| 14 Jan 2026 | STEP3-VL-10B technical report | 63.2 | BeyondAIME, SeRe (Tables 2 and 3). PaCoRe 74.00 | Capabilities |
| 31 Oct 2025 | LongCat-Flash-Omni technical report | 47.4 | BeyondAIME (avg@10) (Table 13) | Capabilities |
| 23 Sep 2025 | LongCat-Flash-Thinking technical report | 69.5 | BeyondAIME(Mean@10) (Table 2) | Capabilities |
| 1 Sep 2025 | LongCat-Flash-Chat technical report | 43 | BeyondAIME (avg@10) (Table 3) | Capabilities |
| 10 Apr 2025 | Seed-Thinking-v1.5 technical report | 48 | Beyond AIME, mean of 32 samples (Table 2) | Capabilities |
More in Math
- IMO-AnswerBench In use 400 olympiad problems with short verifiable answers, robustified against memorization. Scored by answer match. GLM-5.2 91%
- AIME 2026 In use Models solve the 30 integer-answer problems from the 2026 AIME I and II math competitions. Scored by exact answer match. GPT-5.5 100%
- HMMT 2026 In use Models solve the 33 final-answer problems from the February 2026 Harvard-MIT Mathematics Tournament. Scored by exact answer match. GPT-5.5 98.5%
- MathArena (Apex, IMO, USAMO) In use Evaluates models on newly released math competitions after release. Final-answer or human/LLM-judged proof scores. Claude Opus 4.8 81.3%
- PolyMath In use Models solve 9,000 parallel math problems in 18 languages across four difficulty levels, from school math to frontier math. Final answers are checked.
Qwen3.7-Max 86.5% - USAMO 2026 In use Models write full proofs for the six problems of the 2026 USA Mathematical Olympiad. Judges grade each proof from 0 to 7 against rubrics. GPT-5.5 98.2%