GPQA DiamondVerified
91.1%
GPQA Diamond (AA run)
Models / xAI
Grok 4.20 Reasoning
Grok 4 · Released 2026-03-05
xAI's March 2026 reasoning Grok. Reasoning-first variant of the Grok 4 family.
xAI reasoning route when Grok 4.3 is too expensive or too general.
Composite58
Context256K
Input / 1M—
Output / 1M—
Knowledge cutoff2026-01-01
Output speed90 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
66/100 avg
- Reasoning
- 91
- Agentic
- 38
- Long context
- 69
3 tested benchmark families
Benchmark placements
Per-benchmark positions
Where Grok 4.20 Reasoning places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| GPQA Diamond | Reasoning | #36/ 184 | 91 | Artificial Analysis | 2026-09-28 |
| Humanity's Last Exam | Reasoning | #64/ 186 | 35 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #68/ 195 | Artificial Analysis | 2026-10-12 | |
| Terminal-Bench | Agentic | #108/ 167 | 38 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #131/ 184 | 69 | Artificial Analysis | 2026-09-28 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Grok 4.20 Reasoning shows up, it's highlighted.
ReasoningGrok 4.20 Reasoning · 91
- 196
- 295
- 395
Math
- 152
- 252
- 350
AgenticGrok 4.20 Reasoning · 38
- 191
- 289
- 389
Long contextGrok 4.20 Reasoning · 69
- 189
- 286
- 385
Newest receipts
5 recentLast 120 days
Benchmark rows added to the public ledger for Grok 4.20 Reasoning in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 90 t/s | Artificial Analysis | 10 |
| AA-LCR | Long context | 69% | Artificial Analysis | 24 |
| GPQA Diamond | Reasoning | 91.1% | Artificial Analysis | 24 |
| Humanity's Last Exam | Reasoning | 34.5% | Artificial Analysis | 24 |
| Terminal-Bench | Agentic | 37.88% | Artificial Analysis | 24 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Grok 4.20 Reasoning sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Grok 4.20 Reasoning. Scores link to the original source; gaps mean no public row exists yet.
5 of 13 catalog benchmarks have a sourced row for Grok 4.20 Reasoning.
Reasoning
Humanity's Last ExamVerified
34.5%
HLE (AA run)
Agentic
Terminal-BenchVerified
37.88%
Terminal-Bench Hard (AA run)
Long context
AA-LCRVerified
69%
AA-LCR (AA run)
Performance
AA output speedVerified
90 t/s
Median output tokens/s (1k prompt, default provider)
Sources
Where the numbers come from
- xAI docs· checked 2026-06-09
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Grok 4.20 Reasoning?
Grok 4.20 Reasoning is a xAI model: xAI's March 2026 reasoning Grok. Reasoning-first variant of the Grok 4 family.
What is Grok 4.20 Reasoning for?
xAI reasoning route when Grok 4.3 is too expensive or too general.
What does Grok 4.20 Reasoning cost?
Grok 4.20 Reasoning lists — input and — output per 1M tokens. Pricing last checked 2026-06-09.
How large is the Grok 4.20 Reasoning context window?
Grok 4.20 Reasoning publishes a 256K token context window.
What is the VerdictPal status of Grok 4.20 Reasoning?
Grok 4.20 Reasoning carries Solid status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.