grok-3: benchmark scores, price and ranking

grok-3 from xAI ranks #508 of 743 LLMs on the LLMs Tiger aggregate benchmark score (46.5), based on 4 benchmark results captured up to 2026-10-13.

Organisation: xAI
Type: Frontier model
Open weights: No
Released: 2025-02-19 (source: curated)
Cheapest API price: $3.00 in / $15.00 out per million tokens (via azure_ai, observed 2026-10-13T20:47Z)
Aggregate rank: #508 of 743 (score 46.5)

Relative to other models, grok-3 is strongest on Epoch ECI (#59 of 204) and weakest on ARC-AGI-2 (#83 of 87).

Across the 4 benchmarks where at least ten models are scored, grok-3 is in the top 10% on 0 and in the top half on 1.

Its cheapest listed input price, $3.00 per million tokens, is lower than that of 11% of the 427 models with a listed price.

grok-3 benchmark results

grok-3 benchmark scores, ranked against all models measured on each benchmark
Benchmark	Score	Rank	Source	Captured
Epoch ECI	138	#59 of 204	Epoch AI Benchmarking Hub	2026-10-13
SimpleBench	36.1%	#37 of 61	Epoch AI Benchmarking Hub	2026-10-13
WeirdML	37.2%	#73 of 109	Epoch AI Benchmarking Hub	2026-10-13
ARC-AGI-2	0.0%	#83 of 87	ARC Prize official leaderboard	2026-10-13

About grok-3

grok-3 is a frontier model from xAI. It was released on 2025-02-19, according to the developer's announcement. The cheapest API listing we track charges $3.00 per million input tokens and $15.00 per million output tokens, through azure_ai. We hold 4 benchmark results for it across 2 areas, and none is in the top tenth of its board.

Where grok-3 stands

On our aggregate score, grok-3 is #508 of 743 ranked models, in the lower half. Its score is 46.5; the scale is centred on 50, so it sits 3.5 points below the typical model. That beats 32% of the ranked field. Among the 26 ranked xAI models it is #23. The next one up is grok-4-1-fast-reasoning at 46.7. The next one down is grok-build-0.1 at 45.4.

grok-3, benchmark by benchmark

Epoch ECI puts grok-3 at 138, #59 of 204. That beats 71% of the models tested, but 18 points short of Qwen 3.8 Max (156). The results either side are Qwen3-32B (139) and qwen3-14b (138). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so grok-3 is 11 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

grok-3 lands below the middle on SimpleBench: 36.1%, #37 of 61, with 59% of the field scoring higher. Claude Fable 5 (max) leads at 81.9%; the closest result above is o4-mini (high) at 38.7%. It covers trick questions that test common sense. The board median is 41.4%, so grok-3 is 5.3 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

grok-3 lands below the middle on WeirdML: 37.2%, #73 of 109, with 66% of the field scoring higher. GPT-6 Astra (max) leads at 93.3%; the closest result above is qwen3-235b-a22b at 37.3%. It covers unusual machine-learning tasks. The board median is 41.2%, so grok-3 is 3.9 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.

ARC-AGI-2 is a weak spot. grok-3 scores 0.0%, #83 of 87, behind 94% of the models measured. The leader, GPT-6 Astra (max), reaches 95.0%. It covers abstract reasoning, ARC Prize public eval set. The board median is 22.8%, so grok-3 is 22.8 points below it. Source: ARC Prize official leaderboard, captured 2026-10-13.

Strengths and weaknesses

Reasoning, maths and science: weak, ahead of 22% of the field on average across 2 benchmarks. Its best board here is SimpleBench (#37 of 61) and its weakest is ARC-AGI-2 (#83 of 87). General knowledge and chat: solid, ahead of 52% of the field on average across 2 benchmarks. Its best board here is Epoch ECI (#59 of 204) and its weakest is WeirdML (#73 of 109).

Price and value

The median input price across the 427 priced models is $0.500 per million tokens, so grok-3 costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), Qwen2.5-Coder-3B-Instruct ($0.010, 46.5). Similar scores cost more on gpt-5.5-xhigh (codex-harness) ($5.00) and GPT-4 Turbo Preview (Nov 2023) ($10.00).

Release history

The xAI model released before it was grok-2-1212 on 2024-12-12, which scores 43.4. The next one was Grok 3 mini on 2025-02-19, at 49.4.

Reading these numbers

The 4 scores on this page come from 2 sources: Epoch AI Benchmarking Hub (3), ARC Prize official leaderboard (1). 1 of them come from an official leaderboard and 3 from datasets or third-party runs. With 4 results the aggregate rank can move a lot when one new score arrives.

Models ranked nearby

#505 DeepSeek-V2 (46.6)
#506 qwen3.6-plus (46.6)
#507 Phi-3.5-mini-instruct (46.5)
#509 Mellum2-12B-A2.5B-Instruct-SFT (46.5)
#510 Qwen2.5-Coder-3B-Instruct (46.5)
#511 amazon.nova-pro-v1:0 (46.5)