xAI · Sep 21, 2026
xAI introduces Grok 4.7 for coding and knowledge work
Markdown · JSON · Original source
On September 21, 2026, xAI introduced Grok 4.7 as its most capable model for coding and knowledge work. xAI says the model uses a larger base model and a longer reinforcement learning run focused on harder tasks. It says Grok 4.7 is better at checking its work and managing longer context. The announcement reports benchmark results across coding, engineering, office, legal, clinical, and safety tasks. On CursorBench 4.0, which xAI says stresses longer-running coding tasks, Grok 4.7 scored 46.3%. xAI calls its price-performance “at the frontier.” These benchmark results describe tests, not demonstrated performance in everyday professional work. xAI says Grok 4.7 has a new safeguard stack and reports results on its safety benchmarks. The company says the model is available through several products and services. Its API pricing starts at $2 per million input tokens and $6 per million output tokens. How well do Grok 4.7’s benchmark results and safeguards hold up in independent evaluations and real-world use?
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
xAI · Article
Source details
Original source date: Sep 21, 2026 (2026-09-21T00:00:00Z).
Date basis: Date stated on the publisher’s page.
Added to Escape Meridian: 2026-10-29T16:53:41.624Z. First observed: 2026-10-18T02:43:41.439Z. Retrieved: 2026-10-23T20:05:37.865Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
The announcement presents xAI’s own descriptions and benchmark results; it provides no independent evaluation or evidence of performance in everyday professional use. Text as retrieved on October 23, 2026; originally published September 21, 2026.
Open question
How well do Grok 4.7’s benchmark results and safeguards hold up in independent evaluations and real-world use?
Source-stated
xAI describes Grok 4.7 as its most capable model for coding and knowledge work and says it works longer on difficult tasks and checks its work more carefully.
Grok 4.7 is our most capable model for coding and knowledge work. It works longer on difficult tasks, checks its own work more carefully
Source-stated
xAI says Grok 4.7 uses a larger base model than Grok 4.6 and was trained with a longer reinforcement learning run on harder tasks.
It was trained with a longer reinforcement learning run on a harder mix of tasks
Source-stated
xAI reports that Grok 4.7 scored 46.3% on CursorBench 4.0.
Software engineeringCursorBench 4.0 46.3%
Source-stated
xAI says Grok 4.7 was built with a new safeguard stack and reports a 62.4% score on LatchBio’s biosafety benchmark.
topping LatchBio’s biosafety benchmark at 62.4%.
Source-stated
xAI says Grok 4.7 is available through several services and gives starting API token prices.
The model is priced starting at $2 per million input tokens and $6 per million output tokens.
Source and contributor context
xAI · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Reinforcement learning
Training by trial and error, in which a system is rewarded for the outcomes its designers want.
Benchmark
A defined set of tests used to compare systems. Results depend on the tasks and measures selected.
API
An application programming interface: a defined way for software to request data or actions from another system.
Model tokens
Small chunks of text, often parts of words, that a language model reads and writes; usage and prices are often counted in tokens.
xAI
xAI develops the Grok family of AI models and publishes its own release claims and evaluations.
Grok
Grok is a family of AI models from xAI, with capability claims and evaluations published by its developer.
Editorial provenance
Version 1. Publication ID: 25070eff-f36e-6f1d-12d8-0df1d7aa3009.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-10-29T16:52:38.058Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: 8ab2d5ea3919866ff78e663cc00d55bccf71e82c1cbad83f5b0b6fae1e434be8. Source text hash: d396cc2c94bde6b89f2194965208809b820c37172fbb9440f938675705f0847c.
Revision and source-check history
Version 1 · 2026-10-29T16:53:41.624Z
xAI introduces Grok 4.7 for coding and knowledge work
On September 21, 2026, xAI introduced Grok 4.7 as its most capable model for coding and knowledge work. xAI says the model uses a larger base model and a longer reinforcement learning run focused on harder tasks. It says Grok 4.7 is better at checking its work and managing longer context. The announcement reports benchmark results across coding, engineering, office, legal, clinical, and safety tasks. On CursorBench 4.0, which xAI says stresses longer-running coding tasks, Grok 4.7 scored 46.3%. xAI calls its price-performance “at the frontier.” These benchmark results describe tests, not demonstrated performance in everyday professional work. xAI says Grok 4.7 has a new safeguard stack and reports results on its safety benchmarks. The company says the model is available through several products and services. Its API pricing starts at $2 per million input tokens and $6 per million output tokens. How well do Grok 4.7’s benchmark results and safeguards hold up in independent evaluations and real-world use?
Original source
xAI · Article
Source details
Original source date: Sep 21, 2026 (2026-09-21T00:00:00Z).
Date basis: Date stated on the publisher’s page.
Added to Escape Meridian: 2026-10-29T16:53:41.624Z. First observed: 2026-10-18T02:43:41.439Z. Retrieved: 2026-10-23T20:05:37.865Z.
Limits
The announcement presents xAI’s own descriptions and benchmark results; it provides no independent evaluation or evidence of performance in everyday professional use. Text as retrieved on October 23, 2026; originally published September 21, 2026.
Open question
How well do Grok 4.7’s benchmark results and safeguards hold up in independent evaluations and real-world use?
Source-stated
xAI describes Grok 4.7 as its most capable model for coding and knowledge work and says it works longer on difficult tasks and checks its work more carefully.
Grok 4.7 is our most capable model for coding and knowledge work. It works longer on difficult tasks, checks its own work more carefully
Source-stated
xAI says Grok 4.7 uses a larger base model than Grok 4.6 and was trained with a longer reinforcement learning run on harder tasks.
It was trained with a longer reinforcement learning run on a harder mix of tasks
Source-stated
xAI reports that Grok 4.7 scored 46.3% on CursorBench 4.0.
Software engineeringCursorBench 4.0 46.3%
Source-stated
xAI says Grok 4.7 was built with a new safeguard stack and reports a 62.4% score on LatchBio’s biosafety benchmark.
topping LatchBio’s biosafety benchmark at 62.4%.
Source-stated
xAI says Grok 4.7 is available through several services and gives starting API token prices.
The model is priced starting at $2 per million input tokens and $6 per million output tokens.
Source and contributor context
xAI · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Reinforcement learning
Training by trial and error, in which a system is rewarded for the outcomes its designers want.
Benchmark
A defined set of tests used to compare systems. Results depend on the tasks and measures selected.
API
An application programming interface: a defined way for software to request data or actions from another system.
Model tokens
Small chunks of text, often parts of words, that a language model reads and writes; usage and prices are often counted in tokens.
xAI
xAI develops the Grok family of AI models and publishes its own release claims and evaluations.
Grok
Grok is a family of AI models from xAI, with capability claims and evaluations published by its developer.
Editorial provenance
Version 1. Publication ID: 25070eff-f36e-6f1d-12d8-0df1d7aa3009.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: passed by anthropic/claude-opus-5.5 on 2026-10-29T16:52:38.058Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: 8ab2d5ea3919866ff78e663cc00d55bccf71e82c1cbad83f5b0b6fae1e434be8. Source text hash: d396cc2c94bde6b89f2194965208809b820c37172fbb9440f938675705f0847c.
passed · 2026-10-29T16:52:38.058Z · anthropic/claude-opus-5.5
Publication: 25070eff-f36e-6f1d-12d8-0df1d7aa3009
attribution — pass: The reading attributes the capability descriptions, benchmark results, safeguard claims, availability and pricing to xAI, the publisher of the announcement. It frames price-performance as xAI's own characterization.
Grok 4.7 is our most capable model for coding and knowledge work.
summary — pass: The title, summary and question match the announcement: a new model, its training changes, benchmark results, safeguards and pricing. Terms such as reinforcement learning, base model, context and tokens are not explained, but a general reader can still follow the central development. The summary correctly notes that benchmarks are tests rather than demonstrated real-world performance. The question raises a genuine unresolved issue that follows from the evidence.
Grok 4.7 uses a new, larger base model compared to Grok 4.6.
date — pass: The reading's date of September 21, 2026 matches the UTC calendar day of evidence.originalPublished (2026-09-21T00:00:00Z). The evidence kind is publisher-page, so no archive-copy disclosure is required. The reading contains no hindsight framing and no later events, and the retrieval date is disclosed in the limitations.
support — pass: Each claim is supported by the quoted source text. The 46.3% CursorBench figure appears in the column the table labels Grok 4.7 xHigh; the reading omits that setting, but this is a minor imprecision rather than a contradiction. The reading does not overstate benefits and frames real-world reliability as unknown.
On CursorBench 4.0, which stresses longer-running coding tasks, Grok 4.7 is at the frontier in price-performance.
media — pass: No media is selected and no verified media candidates were supplied, so no media identity or relevance issue arises.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.