xAI Proprietary Released Mar 9, 2026 Updated
Grok-4.20 Beta Reasoning
ToolsGuide Scores, rankings, test results and price.
Model summary
Summary
Grok-4.20 Beta Reasoning is xAI's model, released Mar 9, 2026. It's ranked in 3 categories and does best at text: #49 of 416, with a ToolsGuide Score of 494 against a median of 235.
It also ranks #51 of 167 for vision (455), #103 of 147 for coding (266).
Its text and vision and coding score uses votes only, because it has published fewer than two of the tests used for those categories.
It has published results on 4 tests.
Specifications
- Creator
- xAI
- Licence
- Proprietary
- Released
- Mar 9, 2026
- Context window
- –
- Input price
- –
- Output price
- –
- Tests reported
- 4
- Settings ranked
- grok-4.20-beta-0309-reasoning, Grok 4.20 Beta (Reasoning, 0309)
Model comparison
Against the ten best text models.
Text score
ToolsGuide Score out of 1,000 · Higher is better
Vision score
ToolsGuide Score out of 1,000 · Higher is better
Input price
USD per 1M input tokens · Lower is better
Leaderboards
Where Grok-4.20 Beta Reasoning sits among the top 20 in each category.
Text
#49 of 416 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model
Vision
#51 of 167 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model
Coding
#103 of 147 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model
By task
Scored from votes on each kind of prompt alone, on the same 1,000-point scale.
Against its nearest rivals
Text prompts by kind · ToolsGuide Score from votes · hover for values
- Grok-4.20 Beta Reasoning
-
gpt-5.2-high - grok-4.20-multi-agent-beta-0309
Text
grok-4.20-beta-0309-reasoning · hover a bar for its rank
Vision
Grok 4.20 Beta (Reasoning, 0309) · hover a bar for its rank
Coding
Grok 4.20 Beta (Reasoning, 0309) · hover a bar for its rank
Test results
Every test Grok-4.20 Beta Reasoning has a published result on, out of 100%, with its rank among the models that took it.
MRCR v2 (8-needle)
#19 / 32MRCR v2 (8-needle) is a variant of the Multi-Round Coreference Resolution benchmark that includes 8 needle items to retrieve from long contexts. This tests models' ability to simultaneously track and reason about multiple pieces of information across extended conversations.
32.9%
Verified · In the text score
Finance Agent v2
#27 / 28Finance Agent v2 is an agentic financial-analysis benchmark from Vals that evaluates models on real-world finance workflows, measuring their ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses.
28.5%
Self-reported
mrcr-v2 (8needle_32768_65536)
#2 / 1032.9%
Unverified
mrcr-v2 (8needle_4096_8192)
#2 / 1082.7%
Unverified