grok-3-mini-beta (high): benchmark scores, price and ranking

grok-3-mini-beta (high) from xAI ranks #184 of 743 LLMs on the LLMs Tiger aggregate benchmark score (50.5), based on 3 benchmark results captured up to 2026-10-13.

Organisation: xAI
Type: Preview release
Open weights: No
Aggregate rank: #184 of 743 (score 50.5)

Relative to other models, grok-3-mini-beta (high) is strongest on OTIS Mock AIME (#69 of 189) and weakest on WeirdML (#52 of 109).

Across the 3 benchmarks where at least ten models are scored, grok-3-mini-beta (high) is in the top 10% on 0 and in the top half on 3.

grok-3-mini-beta (high) benchmark results

grok-3-mini-beta (high) benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
OTIS Mock AIME	77.8%	#69 of 189	Epoch AI Benchmarking Hub	2026-10-13
GPQA Diamond	75.5%	#128 of 275	Epoch AI Benchmarking Hub	2026-10-13
WeirdML	42.6%	#52 of 109	Epoch AI Benchmarking Hub	2026-10-13

About grok-3-mini-beta (high)

grok-3-mini-beta (high) is a preview release from xAI. No API price is listed for it yet. We hold 3 benchmark results for it across 2 areas, and none is in the top tenth of its board.

Where grok-3-mini-beta (high) stands

On our aggregate score, grok-3-mini-beta (high) is #184 of 743 ranked models, in the top quarter. Its score is 50.5; the scale is centred on 50, so it sits 0.5 points above the typical model. That beats 75% of the ranked field. Among the 26 ranked xAI models it is #8. The next one up is grok-4-1-fast-non-reasoning at 51.0. The next one down is grok-3-beta at 50.0.

grok-3-mini-beta (high), benchmark by benchmark

OTIS Mock AIME puts grok-3-mini-beta (high) at 77.8%, #69 of 189. That beats 63% of the models tested, but 22.2 points short of Claude Fable 5.1 (max) (100.0%). The results either side are Claude Sonnet 4.5 (59k thinking) (77.8%) and o3-mini (high) (76.9%). It covers competition maths (AIME 2024-25 style). The board median is 57.8%, so grok-3-mini-beta (high) is 20.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

On GPQA Diamond, grok-3-mini-beta (high) reaches 75.5%. It ranks #128 of 275, ahead of 53% of the field, and 20.3 points behind GPT-6 Astra (max) at 95.8%. The results either side are NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 (75.6%) and NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 (75.4%). It covers expert-level science questions. The board median is 71.7%, so grok-3-mini-beta (high) is 3.8 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

grok-3-mini-beta (high) manages 42.6% on WeirdML, good for #52 of 109. GPT-6 Astra (max) leads at 93.3%, 50.7 points further up, and 52% of models sit below. The results either side are Grok 3 mini (42.6%) and gemini-2.5-flash-preview-09-2025_16k (41.9%). It covers unusual machine-learning tasks. The board median is 41.2%, so grok-3-mini-beta (high) is 1.4 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

Strengths and weaknesses

Reasoning, maths and science: solid, ahead of 58% of the field on average across 2 benchmarks. Its best board here is OTIS Mock AIME (#69 of 189) and its weakest is GPQA Diamond (#128 of 275).

Reading these numbers

The 3 scores on this page come from 1 source: Epoch AI Benchmarking Hub (3). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 3 other reasoning-effort settings of this model, which redirect here. This page shows the highest. With 3 results the aggregate rank can move a lot when one new score arrives.

grok-3-mini-beta (high) is also listed at 3 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.

Models ranked nearby

#181 NVIDIA-Nemotron-3-Super-120B-A12B-FP8 (50.5)
#182 Claude 3.7 Sonnet w/ Review Heavy (50.5)
#183 GPT-5-codex (high) (50.5)
#185 DeepSeek Chat V3 (prev) (50.5)
#186 gemini-2.5-flash-preview-05-20 (50.5)
#187 Kimi K2 Thinking Turbo (50.5)