Grok 4.20

Grok 4.20 is a reasoning model from xAI. 16 benchmarks count toward its score, in 6 categories.

availableShows if the model has enough results for an index.
IndexOverall score out of 100.52.5 ±5.0
CoverageShare of the index weight with results.90%
SpeedOutput tokens per second.58/s
Input / 1MUS dollars per 1M input tokens.$1.25
Output / 1MUS dollars per 1M output tokens.$2.5
ContextMaximum tokens in one request.2M
EloLMArena rating and rank.N/A

The index is a score out of 100. The ± range shows how much it can change.

CapabilitiesScore per category, out of 100.

Out of 100
AgenticMulti-step tasks with tools.
47.2
CodingCode writing and repair.
56.6
ReasoningLogic problems and puzzles.
54.6
MultimodalTasks with images and text.
45.5
KnowledgeFacts and expert knowledge.
55.6
MultilingualTasks in many languages.
N/A
InstructionTasks with strict rules in the prompt.
N/A
MathMath problems.
58.7

Results

16 counted
BenchmarkThe test name.CategoryThe capability that the test measures.ResultThe score from the publisher.IndexThis result as a score out of 100.RunThe settings of the run.DateDate of the result.Published byThe source of the result.
OTIS Mock AIME 2024-2025Math92.2%63.1Epoch AI
GPQA diamondKnowledge89.3%60.7Epoch AI
GPQA DiamondKnowledge88.5%59.9David Rein et al.
Software Engineering Benchmark VerifiedCoding76.7%59.1Carlos E. Jimenez et al.
Massive Multi-discipline Multimodal Understanding ProMultimodal75.2%49.7MMMU-Pro authors
LiveCodeBench ProCoding74.2%LiveCodeBench Pro authors
MedXpertQA MultimodalMultimodal65.8%Meta AI
DeepSearchQAAgentic62.8%47.8Meta AI
CharXiv ReasoningMultimodal60.9%35.4CharXiv authors
SimpleVQAMultimodal57.4%47.4Z.AI
ERQAMultimodal54.1%49.4Qwen
SWE-bench ProCoding51.8%54.1Xiang Deng et al.
MedXpertQA TextKnowledge50.2%Meta AI
FrontierMath-Tiers-1-3-v2-PrivateMath44.9%56.7Epoch AI
Gert Labs Composite Game BenchmarkAgentic38.4%46.6Gert Labs
Humanity's Last Exam without toolsKnowledge31.6%55.6OpenAI
SimpleQA VerifiedKnowledge30.2%49.5Epoch AI
Chess PuzzlesReasoning24.0%54.6Epoch AI
HealthBench HardKnowledge20.3%57.1Meta AI
FrontierMath-Tier-4-v2-PrivateMath17.1%56.1Epoch AI

16 benchmarks count, from 17 of 20 results. A grey row does not count. Too few models took that benchmark.

Sources

BenchLM benchmark aggregationUsed with attribution; per-benchmark results credited to their original authorsOpenRouter, collected directlyNo licence statedEpoch AI, collected directlyCC BY — free to use and redistribute with attribution

Same level, lower price

Qwen3.8-Flash-Next66.4 · Freedots3-note Preview65.5 · FreeApodex 1.1 Mini59.1 · Free

More from xAI

Grok 4.671.8Grok 4.770.1Grok 4.566.0Grok 4.356.0Grok 4.20 (Reasoning)56.9