#156
XA
Grok 1.5
xAI
19.3 LLMBoard
xAI model product
Grok 4.20 Beta Reasoning is xAI's reasoning-optimized variant of the Grok 4.20 model family. Released as a beta on March 9, 2026, it uses thinking tokens for enhanced chain-of-thought reasoning, featuring a 2 million token context window for complex problem-solving tasks.
Updated Aug 10, 2026. Default version: Grok-4.20 Beta Reasoning
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
No published version under this unique model currently has enough benchmark coverage to calculate a score.
Published benchmark records for the scored version currently unavailable.
No published benchmark result is linked to the scored version.
Preference and agent-evaluation signals from published Arena datasets.
| vision | entity recognition | 9 | 1282.2 | 147 | N/A | |
| vision style control | entity recognition | 9 | 1247.4 | 147 | N/A | |
| text factuality | korean | 12 | 1410.0 | 544 | N/A | |
| text factuality | french | 14 | 1495.9 | 1,583 | N/A | |
| text factuality | german | 17 | 1423.6 | 461 | N/A | |
| text style control | french | 17 | 1503.8 | 2,077 | N/A | |
| vision | humor | 17 | 1285.7 | 856 | N/A | |
| vision style control | humor | 17 | 1269.4 | 856 | N/A | |
| vision style control | creative writing vision | 18 | 1273.0 | 1,402 | N/A | |
| text factuality | japanese | 20 | 1420.9 | 442 | N/A | |
| vision | creative writing vision | 20 | 1284.8 | 1,402 | N/A | |
| text style control | russian | 24 | 1480.2 | 6,444 | N/A | |
| text factuality | math | 25 | 1458.2 | 2,568 | N/A | |
| text style control | industry entertainment and sports and media | 25 | 1447.7 | 13,405 | N/A | |
| text factuality | polish | 26 | 1428.3 | 762 | N/A | |
| text factuality | russian | 27 | 1471.0 | 5,038 | N/A | |
| text | french | 28 | 1476.8 | 2,077 | N/A | |
| text factuality | spanish | 28 | 1446.1 | 1,357 | N/A | |
| text style control | industry medicine and healthcare | 28 | 1487.6 | 4,488 | N/A | |
| text style control | korean | 28 | 1431.6 | 1,019 | N/A | |
| text style control | non english | 28 | 1461.8 | 32,810 | N/A | |
| text factuality | industry medicine and healthcare | 29 | 1488.9 | 3,515 | N/A | |
| document style control | overall | 30 | 1439.7 | 18,666 | N/A | |
| text | russian | 31 | 1459.5 | 6,444 | N/A | |
| text style control | overall | 31 | 1471.9 | 61,639 | N/A | |
| text style control | english | 31 | 1478.3 | 28,829 | N/A | |
| text style control | exclude ties | 31 | 1479.9 | 46,942 | N/A | |
| text style control | industry legal and government | 31 | 1478.1 | 4,843 | N/A | |
| text style control | polish | 31 | 1471.7 | 1,209 | N/A | |
| text style control | spanish | 31 | 1465.0 | 1,877 | N/A | |
| document | overall | 33 | 1413.9 | 18,666 | N/A | |
| text style control | multi turn | 33 | 1480.0 | 11,264 | N/A | |
| text style control | industry business and management and financial operations | 34 | 1466.2 | 12,374 | N/A | |
| text style control | industry life and physical and social science | 34 | 1486.8 | 9,971 | N/A | |
| vision style control | chinese | 34 | 1296.5 | 1,350 | N/A | |
| text factuality | english | 35 | 1466.4 | 28,683 | N/A | |
| text style control | creative writing | 35 | 1446.4 | 10,342 | N/A | |
| text | overall | 36 | 1451.2 | 61,639 | N/A | |
| text | english | 36 | 1457.4 | 28,829 | N/A | |
| text | industry entertainment and sports and media | 36 | 1428.5 | 13,405 | N/A | |
| text | polish | 36 | 1449.8 | 1,209 | N/A | |
| text style control | hard prompts english | 36 | 1490.1 | 19,525 | N/A | |
| vision style control | overall | 36 | 1255.0 | 24,509 | N/A | |
| text | exclude ties | 37 | 1449.9 | 46,942 | N/A | |
| text factuality | industry entertainment and sports and media | 37 | 1434.4 | 11,910 | N/A | |
| text style control | industry software and it services | 37 | 1502.7 | 24,159 | N/A | |
| vision style control | english | 37 | 1252.1 | 10,131 | N/A | |
| vision style control | ocr | 37 | 1263.8 | 17,544 | N/A | |
| text | korean | 38 | 1414.4 | 1,019 | N/A | |
| text | non english | 38 | 1441.6 | 32,810 | N/A | |
| text factuality | multi turn | 38 | 1466.2 | 9,396 | N/A | |
| text style control | hard prompts | 38 | 1488.6 | 40,066 | N/A | |
| vision | overall | 38 | 1262.2 | 24,509 | N/A | |
| vision | chinese | 38 | 1302.7 | 1,350 | N/A | |
| text factuality | overall | 39 | 1455.6 | 61,547 | N/A | |
| text factuality | creative writing | 39 | 1440.0 | 8,618 | N/A | |
| text factuality | industry writing and literature and language | 39 | 1445.2 | 13,360 | N/A | |
| vision | english | 39 | 1262.3 | 10,131 | N/A | |
| text factuality | exclude ties | 40 | 1458.2 | 46,868 | N/A | |
| text factuality | non english | 40 | 1442.1 | 32,710 | N/A | |
| text style control | coding | 40 | 1510.6 | 17,074 | N/A | |
| text style control | chinese | 41 | 1502.8 | 3,399 | N/A | |
| text style control | industry writing and literature and language | 41 | 1450.3 | 15,054 | N/A | |
| vision | ocr | 41 | 1264.4 | 17,544 | N/A | |
| text | multi turn | 42 | 1456.9 | 11,264 | N/A | |
| text factuality | chinese | 42 | 1489.8 | 2,689 | N/A | |
| text factuality | industry legal and government | 42 | 1467.8 | 3,832 | N/A | |
| text style control | math | 42 | 1467.3 | 3,304 | N/A | |
| vision style control | diagram | 42 | 1272.8 | 6,535 | N/A | |
| text factuality | industry life and physical and social science | 43 | 1475.9 | 8,312 | N/A | |
| text | industry life and physical and social science | 44 | 1460.7 | 9,971 | N/A | |
| vision | diagram | 44 | 1265.2 | 6,535 | N/A | |
| text | creative writing | 45 | 1432.6 | 10,342 | N/A | |
| text | industry writing and literature and language | 45 | 1435.6 | 15,054 | N/A | |
| text | math | 45 | 1457.3 | 3,304 | N/A | |
| text factuality | hard prompts | 45 | 1475.0 | 39,956 | N/A | |
| text factuality | hard prompts english | 46 | 1484.0 | 18,141 | N/A | |
| text factuality | industry mathematical | 46 | 1441.6 | 2,530 | N/A | |
| text | industry medicine and healthcare | 47 | 1454.7 | 4,488 | N/A | |
| vision | homework | 47 | 1267.2 | 3,423 | N/A | |
| vision style control | homework | 47 | 1263.4 | 3,423 | N/A | |
| text | industry legal and government | 48 | 1454.2 | 4,843 | N/A | |
| text style control | german | 48 | 1455.5 | 1,010 | N/A | |
| text style control | longer query | 50 | 1465.8 | 26,010 | N/A | |
| text | german | 51 | 1440.7 | 1,010 | N/A | |
| text | spanish | 51 | 1447.1 | 1,877 | N/A | |
| text factuality | industry software and it services | 51 | 1486.2 | 23,346 | N/A | |
| text | hard prompts | 52 | 1451.6 | 40,066 | N/A | |
| text factuality | industry business and management and financial operations | 53 | 1452.7 | 10,464 | N/A | |
| text factuality | longer query | 53 | 1461.3 | 24,545 | N/A | |
| text style control | instruction following | 53 | 1446.4 | 20,501 | N/A | |
| text | japanese | 54 | 1405.5 | 648 | N/A | |
| text factuality | instruction following | 54 | 1443.1 | 19,035 | N/A | |
| text style control | industry mathematical | 54 | 1457.4 | 3,268 | N/A | |
| text style control | japanese | 54 | 1419.1 | 648 | N/A | |
| text | chinese | 55 | 1482.1 | 3,399 | N/A | |
| text | hard prompts english | 55 | 1454.5 | 19,525 | N/A | |
| text | industry business and management and financial operations | 55 | 1432.1 | 12,374 | N/A | |
| text factuality | coding | 57 | 1497.9 | 15,658 | N/A | |
| text | industry software and it services | 58 | 1461.3 | 24,159 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| Vercel AI Gateway | xai/grok-4.20-reasoning-beta | global | $1.25 | $2.5 | 2M | |
| 302.AI | grok-4.20-beta-0309-reasoning | global | $2 | $6 | 2M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| Grok-4.20 Beta Reasoning | N/A | N/A | 2M | 30K | No | Proprietary |
A concise description based on the published model registry.
Grok 4.20 Beta Reasoning is xAI's reasoning-optimized variant of the Grok 4.20 model family. Released as a beta on March 9, 2026, it uses thinking tokens for enhanced chain-of-thought reasoning, featuring a 2 million token context window for complex problem-solving tasks.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Grok 4.20 Beta Reasoning.
Grok 4.20 Beta Reasoning's default version was released on Mar 9, 2026.
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.
Grok 4.20 Beta Reasoning is published under xAI in the model registry.
The default version has a 2M token context window.
No. The default version is not marked as having publicly available weights.
2 published provider offerings are linked to the default version.
No nearby ranked alternatives are currently available.