Grok 3 Mini Reasoning
Grok 3 Mini Reasoning is behind the leaders in long context, instruction following, and reasoning. Too few results yet to rate coding, agentic tasks, safety, math, multimodal tasks, multilingual tasks, or factuality.
Capability profile
Bars show the model's median result as a share of the leading model's, per capability. Select a row to see its results.
Too few results yet to rate Coding, Agentic, Safety, Math, Multimodal, Multilingual or Factuality.
Price
$0.30input$0.50outputper million tokens
From Artificial Analysis · All prices
Evidence
30results on25benchmarks
- 7 independently verified
- 15 aggregator
- 8 vendor-reported
From 7 sources · latest Oct 8, 2026 · How verification works
Grok 3 Mini Reasoning benchmark results
30 results on 25 benchmarks, grouped by capability. Each bar is the result as a share of the capability leader's; every result links to its source.
34.0% behind the leader1 of 3 ranked benchmarks measured
- 55.33Oct 8, 2026
40.7% behind the leader1 of 3 ranked benchmarks measured
- IFBench45.85Oct 8, 2026aa_ifbench
45.3% behind the leader4 of 6 ranked benchmarks measured
- GPQA Diamond79.09Oct 8, 2026gpqa
- Humanity's Last Exam11.01Oct 8, 2026aa_hle
- 0.57Oct 8, 2026
- 0.42May 10, 2026
Show 3 more reasoning resultsHide 3 reasoning results
- GPQA Diamond84.00Oct 7, 2026GPQA
- 79.00Jul 13, 2026
- Humanity's Last Exam11.00Jul 13, 2026HLE (no tools)
0 of 10 ranked benchmarks measured
- Terminal-Bench Hard17.42Oct 8, 2026aa_terminalbench_hard
- SciCode40.63Sep 4, 2026aa_scicode
0 of 7 ranked benchmarks measured
- 0.00Jun 15, 2026
0 of 4 ranked benchmarks measured
- 21.10May 20, 2026
- AA-Omniscience · Accuracy15.12Oct 8, 2026omniscienceAccuracy
- AA-Omniscience · Non-hallucination74.18Oct 8, 2026omniscienceNonHallucination
More results
Benchmarks outside the capability baskets. They are not ranked against a leader.
- 16.50May 10, 2026
- 98.14May 3, 2026
- 51.14May 3, 2026
- 85.90May 3, 2026
- 78.10May 3, 2026
- -6.80Oct 8, 2026
Show 9 more resultsHide 9 results
- 31.15Jun 18, 2026
- AA Intelligence14.62Oct 8, 2026aa_intelligence_index
- 90.80Oct 7, 2026
- AIME 202583.00Jul 13, 2026AIME 2025 (no tools)
- Artificial Analysis Coding Index25.16Jun 18, 2026aa_coding_index
- HMMT 202574.00Jul 13, 2026HMMT 2025 (no tools)
- 80.40Aug 23, 2026
- 70.00Jul 13, 2026
- τ²-Bench Telecom (AA run)90.35Oct 8, 2026aa_tau2
Grok 3 Mini Reasoning: common questions
Who makes Grok 3 Mini Reasoning?
Grok 3 Mini Reasoning is made by SpaceXAI.
When was Grok 3 Mini Reasoning released?
Grok 3 Mini Reasoning was released on Feb 19, 2025, according to Artificial Analysis.
What is Grok 3 Mini Reasoning good at?
Grok 3 Mini Reasoning is behind the leaders in long context, instruction following, and reasoning. Too few results yet to rate coding, agentic tasks, safety, math, multimodal tasks, multilingual tasks, or factuality.
How much does Grok 3 Mini Reasoning cost?
Grok 3 Mini Reasoning costs $0.30 per million input tokens and $0.50 per million output tokens, according to Artificial Analysis. At a mix of three input tokens to one output token, it is cheaper than 67% of the 331 priced models we track.
How many benchmarks has Grok 3 Mini Reasoning been tested on?
We track 30 results for Grok 3 Mini Reasoning on 25 benchmarks from 7 sources, 7 of them independently verified. The latest was recorded on Oct 8, 2026.
About this record
Where Grok 3 Mini Reasoning's numbers come from, and every name it appears under.
- Tracked since
- Apr 27, 2026
- Newest source mention
- Jul 13, 2026
Where the results come from
Verification: 30 scores · 7 independently verified · 15 aggregator-attributed · 8 vendor-reported. How these tiers are assigned
From 7 sources on 6 sites. Artificial Analysis supplies 15 of them; the 7 independently verified results come from 3 sites. Bars are coloured by trust tier.
- artificialanalysis.ai15
- x.ai5
- livecodebench.github.io4
- api.llm-stats.com3
- arcprize.org2
- epoch.ai1
Also known as
How our sources name Grok 3 Mini Reasoning at each reasoning setting.
| Setting | Short form | API id |
|---|---|---|
| low | Grok 3 Mini (Low) | — |
| high | grok 3 mini reasoning (high) | grok-3-mini (high) |