xAI model product
Grok 4.20 Beta Reasoning
20 model family, released on March 9, 2026.
Updated Sep 8, 2026. Default version: Grok-4.20 Beta Reasoning
LLMBoard Score55.4Grok-4.20 Beta Reasoning
Coverage0%1 benchmark families
Context window2MTokens
Official input priceN/AOfficial price unavailable
Grok 4.20 Beta Reasoning Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Grok 4.20 Beta Reasoning Benchmark Results
Benchmark scores for Grok-4.20 Beta Reasoning.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkFinance Agent v2 | Score28.49% | Rank25 | Participants26 | Percentile4.00% | EvidenceB | Evaluated |
| BenchmarkLM Arena Document Style Control | Score1,439.71 rating | Rank26 | Participants33 | Percentile21.88% | EvidenceA | Evaluated |
| BenchmarkLM Arena Document | Score1,413.91 rating | Rank29 | Participants33 | Percentile12.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,471.72 rating | Rank29 | Participants210 | Percentile86.60% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision Style Control | Score1,255.85 rating | Rank30 | Participants107 | Percentile72.64% | EvidenceA | Evaluated |
| BenchmarkLM Arena Vision | Score1,263.10 rating | Rank32 | Participants107 | Percentile70.75% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,450.95 rating | Rank34 | Participants210 | Percentile84.21% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,455.26 rating | Rank37 | Participants121 | Percentile70.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Webdev | Score1,373.66 rating | Rank67 | Participants97 | Percentile31.25% | EvidenceA | Evaluated |
Grok 4.20 Beta Reasoning Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision | Categoryentity recognition | Rank08 | Rating / score1,279.81 | Votes145 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank08 | Rating / score1,244.24 | Votes145 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryfrench | Rank13 | Rating / score1,497.83 | Votes1,650 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank15 | Rating / score1,503.10 | Votes2,116 | ObservationsN/A | Result date |
| Arenavision style control | Categorycreative writing vision | Rank15 | Rating / score1,265.87 | Votes1,584 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryjapanese | Rank17 | Rating / score1,421.83 | Votes464 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank17 | Rating / score1,412.76 | Votes625 | ObservationsN/A | Result date |
| Arenavision | Categorycreative writing vision | Rank17 | Rating / score1,277.81 | Votes1,584 | ObservationsN/A | Result date |
| Arenavision | Categoryhumor | Rank19 | Rating / score1,279.55 | Votes921 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank20 | Rating / score1,263.48 | Votes921 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryrussian | Rank22 | Rating / score1,471.28 | Votes5,216 | ObservationsN/A | Result date |
| Arenatext style control | Categoryrussian | Rank22 | Rating / score1,481.41 | Votes6,664 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank23 | Rating / score1,412.55 | Votes567 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank23 | Rating / score1,488.65 | Votes3,588 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank23 | Rating / score1,433.66 | Votes1,029 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank24 | Rating / score1,476.47 | Votes2,116 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry entertainment and sports and media | Rank24 | Rating / score1,448.14 | Votes13,583 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank25 | Rating / score1,487.21 | Votes4,562 | ObservationsN/A | Result date |
| Arenadocument style control | Categoryoverall | Rank26 | Rating / score1,439.71 | Votes18,666 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank26 | Rating / score1,425.50 | Votes823 | ObservationsN/A | Result date |
| Arenatext style control | Categorypolish | Rank26 | Rating / score1,472.14 | Votes1,224 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank27 | Rating / score1,460.30 | Votes6,664 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymath | Rank27 | Rating / score1,458.11 | Votes2,700 | ObservationsN/A | Result date |
| Arenatext style control | Categorymulti turn | Rank27 | Rating / score1,479.92 | Votes11,425 | ObservationsN/A | Result date |
| Arenatext style control | Categorynon english | Rank27 | Rating / score1,461.78 | Votes33,588 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank28 | Rating / score1,416.01 | Votes1,029 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank28 | Rating / score1,450.06 | Votes1,224 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank28 | Rating / score1,477.70 | Votes28,663 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank28 | Rating / score1,462.59 | Votes2,003 | ObservationsN/A | Result date |
| Arenadocument | Categoryoverall | Rank29 | Rating / score1,413.91 | Votes18,666 | ObservationsN/A | Result date |
Grok 4.20 Beta Reasoning Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $1.25 input, $2.5 output per 1M via Vercel AI Gateway
- Tracked offerings
- 2
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderVercel AI Gateway | Provider model IDspacexai/grok-4.20-reasoning-beta | Regionglobal | Input / 1M$1.25 | Output / 1M$2.5 | Context2M | Updated |
| Provider302.AI | Provider model IDgrok-4.20-beta-0309-reasoning | Regionglobal | Input / 1M$2 | Output / 1M$6 | Context2M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Grok 4.20 Beta Reasoning Runtime Performance
Provider-specific output speed and catalog latency for Grok-4.20 Beta Reasoning. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Grok 4.20 Beta Reasoning Specifications
Technical details for the model's default version.
- Version
- Grok-4.20 Beta Reasoning
- Released
- Mar 9, 2026
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 2M
- Max output
- 30K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Grok 4.20 Beta Reasoning Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGrok-4.20 Beta Reasoning | Released | LLMBoard55.41 | ParametersN/A | Context2M | Max output30K | Open weightsNo | LicenseProprietary |
Models similar to Grok 4.20 Beta Reasoning
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
XA
Grok 4
xAI
XA
Grok 3
xAI
XA
Grok 4 Fast
xAI
XA
Grok 4.1 Thinking
xAI
XA
Grok 3 Mini
xAI
XA
Grok 4 Heavy
xAI
What is Grok 4.20 Beta Reasoning?
Key information about Grok 4.20 Beta Reasoning and its available data.
20 Beta Reasoning is a large language model from xAI that uses thinking tokens for chain-of-thought reasoning and has a 2 million token context window.
Data as of 2026-09-08.
FAQ
Common questions about Grok 4.20 Beta Reasoning.
When was Grok 4.20 Beta Reasoning released?
Grok 4.20 Beta Reasoning's default version was released on Mar 9, 2026.
How much does Grok 4.20 Beta Reasoning cost?
No official standard PAYG price is currently available for Grok 4.20 Beta Reasoning. The lowest tracked third-party offer starts at $1.25 input and $2.5 output via Vercel AI Gateway.
Who created Grok 4.20 Beta Reasoning?
Grok 4.20 Beta Reasoning was created by xAI.
What is the context window for Grok 4.20 Beta Reasoning?
The default version has a 2M token context window.
Is Grok 4.20 Beta Reasoning open weight?
No. The default version is not marked as having publicly available weights.