AI co-mathematician: benchmark scores, price and ranking

AI co-mathematician from Google DeepMind ranks #69 of 743 LLMs on the LLMs Tiger aggregate benchmark score (52.3), based on 1 benchmark results captured up to 2026-10-13.

Organisation: Google DeepMind
Type: Frontier model
Open weights: No
Released: 2026-05-08 (source: epoch)
Aggregate rank: #69 of 743 (score 52.3)

Across the 1 benchmark where at least ten models are scored, AI co-mathematician is in the top 10% on 0 and in the top half on 1.

AI co-mathematician benchmark results

AI co-mathematician benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
FrontierMath T4	75.6%	#9 of 50	Epoch AI Benchmarking Hub	2026-10-13

About AI co-mathematician

AI co-mathematician is a frontier model from Google DeepMind. It was released on 2026-05-08, according to Epoch AI. No API price is listed for it yet. We hold 1 benchmark result for it across 1 area, and none is in the top tenth of its board.

Where AI co-mathematician stands

On our aggregate score, AI co-mathematician is #69 of 743 ranked models, in the top quarter. Its score is 52.3; the scale is centred on 50, so it sits 2.3 points above the typical model. That beats 91% of the ranked field. Among the 41 ranked Google DeepMind models it is #5. The next one up is Gemini 3 Deep Think at 52.6. The next one down is Gemini 3 Flash at 51.5.

AI co-mathematician, benchmark by benchmark

FrontierMath T4 puts AI co-mathematician at 75.6%, #9 of 50. That beats 82% of the models tested, but 24.4 points short of GPT-6.1 Sol (max) (100.0%). The results either side are Claude Sonnet 5.5 (max) (80.5%) and Claude Opus 5 (max) (73.2%). It covers research-level maths, tier 4. The board median is 30.5%, so AI co-mathematician is 45.1 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Release history

The Google DeepMind model released before it was Gemma 4 31B IT on 2026-04-02, which scores 47.3. The next one was Gemini 3.7 Flash (high) on 2026-08-13, at 53.9.

Reading these numbers

The 1 score on this page comes from 1 source: Epoch AI Benchmarking Hub (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 1 results the aggregate rank can move a lot when one new score arrives.