Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Grok 4.20

SpaceXAI

Strongest in Mathematics (#18 of 140, on 1 of its 2 boards), weakest in Coding (#147 of 165). Above par in 8 of 14 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of Qwen3 Max and Mimo V2.5 Pro.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Core abilities57.6#19/204
General intelligence60.9#20/204
Reasoning57.8#23/178
Mathematics58#18/140
Reasoning56.4#34/140
Science58.8#34/122
Knowledge55.2#47/138
Academic knowledge56.3#43/123
Professional43.7#134/168
Finance44.2#116/151
Medical43.3#117/140
Legal42.4#121/151
Coding41.1#147/165
Agentic coding40#143/157

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for Grok 4.20, left for the other.

Laguna XS.2won 10 · lost 0
Mistral Smallwon 10 · lost 0
Mistral Medium 3.5won 13 · lost 0
GPT-4o Miniwon 12 · lost 1
Llama 4 Scout Instructwon 11 · lost 1
GPT OSS 20Bwon 10 · lost 1
Mercury 2.5won 9 · lost 1
Jamba 1.6 Miniwon 8 · lost 1
Jamba 1.6 Largewon 8 · lost 1
Mistral Small 3.1won 8 · lost 1
Magistral Small 1.2won 8 · lost 1
Magistral Medium 1.2won 8 · lost 1
GPT-4.1 Nanowon 12 · lost 2
GPT-5 Nanowon 10 · lost 2
Gemini 1.5 Prowon 8 · lost 2
Laguna M.1won 8 · lost 2
Command Awon 10 · lost 3
Mistral Large 3won 10 · lost 4
Grok 4 Fastwon 10 · lost 4
Ling 3.0 Flashwon 7 · lost 3
Claude 3.5 Haikuwon 7 · lost 3
GPT-5.4 Miniwon 9 · lost 4
Gemini 3.1 Flash Litewon 9 · lost 4
Grok 2won 6 · lost 3
Grok 4.1 Fastwon 8 · lost 4
Grok 3won 8 · lost 4
Grok 4.3won 9 · lost 5
GPT-5.4 Nanowon 9 · lost 5
Gemini 2.0 Flashwon 9 · lost 5
GPT OSS 120Bwon 9 · lost 5
Gemini 2.5 Flash Litewon 7 · lost 4
GLM-4.5won 7 · lost 4
Grok 3 Miniwon 8 · lost 5
DeepSeek R1won 8 · lost 5
Qwen3.6 27Bwon 6 · lost 4
GLM-4.7won 7 · lost 5
O3 Miniwon 7 · lost 5
Gemini 3.5 Flash Litewon 8 · lost 6
Claude Haiku 4.5won 8 · lost 6
GLM-5.3 Flashwon 8 · lost 6
GPT-4.1 Miniwon 8 · lost 6
Kimi K2won 8 · lost 6
DeepSeek V3.2won 8 · lost 6
DeepSeek V3won 8 · lost 6
GPT-4.1won 8 · lost 6
MiniMax M2.5won 8 · lost 6
Gemini 2.5 Flashwon 8 · lost 6
Kimi K2 Thinkingwon 8 · lost 6
O4 Miniwon 8 · lost 6
O3won 8 · lost 6
Gemini 2.5 Prowon 8 · lost 6
GPT-5 Miniwon 8 · lost 6
Claude Sonnet 4won 8 · lost 6
Grok 4won 8 · lost 6
Claude Opus 4won 8 · lost 6
GLM-5won 8 · lost 6
MiniMax M2.7won 6 · lost 5
Qwen3.5 Flashwon 6 · lost 5
Mimo V2.5won 6 · lost 5
GLM-4.6won 7 · lost 6
Nemotron 3 Ultrawon 7 · lost 6
Mimo V2.5 Prowon 7 · lost 6
Qwen3 Maxwon 7 · lost 6
Gemma 4 31Bwon 5 · lost 5
GPT-4owon 6 · lost 6
Inkling Smallwon 6 · lost 6
O1won 6 · lost 6
Claude 3.7 Sonnetwon 6 · lost 6
MiniMax M3won 7 · lost 7
GPT-5.6 Lunawon 7 · lost 7
Qwen3.8 27Bwon 7 · lost 7
Claude 3.5 Sonnetwon 5 · lost 6
GPT-6 Lunawon 5 · lost 6
Qwen3.6 Pluswon 6 · lost 8
Inklingwon 6 · lost 8
GLM-5.2won 6 · lost 8
Claude Sonnet 4.6won 6 · lost 8
DeepSeek V4 Flashwon 6 · lost 8
Claude Sonnet 4.5won 6 · lost 8
Claude Opus 4.5won 6 · lost 8
GPT-5.2won 6 · lost 8
GLM-5.3won 6 · lost 8
GPT-5won 6 · lost 8
Kimi K2.5won 6 · lost 8
Claude Opus 4.1won 4 · lost 6
Kimi K2.6won 5 · lost 9
Muse Spark 1.1won 5 · lost 9
Gemini 3.6 Flashwon 5 · lost 9
Qwen3.7 Maxwon 5 · lost 9
DeepSeek V4 Prowon 5 · lost 9
Gemini 3 Flashwon 4 · lost 8
GPT-5.1won 4 · lost 8
GLM-5.1won 4 · lost 8
MiniMax M2.1won 4 · lost 8
Muse Spark 1.2won 4 · lost 9
DeepSeek V4.1 Flashwon 4 · lost 9
Claude Sonnet 5won 4 · lost 10
Gemini 3.5 Flashwon 4 · lost 10
Qwen 3.5 Pluswon 3 · lost 8
Grok 4.7won 3 · lost 8
Mimo V2.6 Prowon 2 · lost 7
Claude Opus 4.6won 3 · lost 11
Gemini 3.8 Flashwon 3 · lost 11
Grok 4.5won 3 · lost 11
Gemini 3.7 Flashwon 3 · lost 11
Grok 4.6won 3 · lost 11
Qwen3.8 Maxwon 3 · lost 11
Gemini 3.1 Prowon 3 · lost 11
DeepSeek V4won 2 · lost 9
Claude Opus 4.8won 2 · lost 12
Claude Opus 4.7won 2 · lost 12
GPT-5.6 Terrawon 2 · lost 12
Claude Opus 5won 2 · lost 12
GPT-5.4won 2 · lost 12
Kimi K3won 2 · lost 12
Muse Spark 1.3won 1 · lost 9
Muse Sparkwon 1 · lost 10
GPT-5.6 Solwon 1 · lost 13
GPT-6 Solwon 0 · lost 11
Claude Opus 5.5won 0 · lost 11
GPT-6 Astrawon 0 · lost 11
Gemini 3 Prowon 0 · lost 14
GPT-5.5won 0 · lost 14
Claude Fable 5.1won 0 · lost 14
Claude Fable 5won 0 · lost 14

§ 3 · Sources

Where the numbers come from

3 publications, 19 figures. Every one links to the page it was read from.

ARC-AGI-2 ↗1 measureread 2026-09-26
ARC-AGI-2 65.1%
Kagi LLM Benchmark ↗1 measureread 2026-09-26
Kagi LLM Benchmark 75%
Vals.ai ↗17 measuresread 2026-09-26
Vals · Legal Research Bench 13.94%Vals · CaseLaw 54.45%Vals · LegalBench 77.74%Vals · Harvey Legal Agent Benchmark 0%Vals · Finance Agent 28.49%Vals · CorpFin 63.67%Vals · TaxEval 74.12%Vals · MortgageTax 45.35%Vals · MedQA 94.55%Vals · MedCode 32.16%Vals · MedScribe 63.41%Vals · SWE-bench Verified 72.2% (Mini-SWE-agent)Vals · Vibe Code Bench 4.06% (OpenHands)Vals · Code Migration 0.31%Vals · GPQA Diamond 88.64%Vals · MMLU Pro 86.25%Vals · AIME 96.46%

Badge

PublicAI Index badge for Grok 4.20[![PublicAI Index](https://publicai.io/model-index/badge?model=grok-4-20)](https://publicai.io/model-index/m/grok-4-20)