Xai

Grok 4.5

Version 4.5 benchmark runs across the shared Superbash visual prompts.

Canonical model record

Current identity, limits, and pricing

Provider source · checked 2026-08-14 ↗

Status: Current
API model ID: grok-4.5
Context: 500K
Max output: Not publicly verified
API price / 1M tokens: $2 input · $6 output

Prompts of 200K tokens or more use the provider's higher long-context rate.

Visual prompt runs

Benchmark runs

Open each generated scene, or compare the same prompt across models.

7 runs

Hogwarts Broom Flight Simulator

Broom-flight scene testing depth, motion cues, castle scale, and fantasy mood.

Open run →Compare this model →

Jabberwock

Dark fantasy encounter testing creature design, forest mood, and narrative staging.

Open run →Compare this model →

Low Poly World

Stylized island build testing composition, color, and low-poly worldbuilding.

Open run →Compare this model →

Office Life

Workplace vignette testing everyday scene logic, objects, and believable office detail.

Open run →Compare this model →

Petri Dish

Microscopic ecosystem testing organic forms, scientific clarity, and cellular detail.

Open run →Compare this model →

Universe Simulator

Cosmic system testing orbital structure, glowing bodies, scale, and simulation readability.

Open run →Compare this model →

Vice City

Neon coastal city testing vehicles, architecture, atmosphere, and dense urban layout.

Open run →Compare this model →

Official benchmark profile

How Grok 4.5 scores beyond our visual tests.

xAI describes Grok 4.5 as a coding, agentic-task, and knowledge-work model. xAI publishes different benchmark rows than OpenAI/Anthropic, so only direct source comparisons are shown.

SpaceXAI sourceJuly 2026Source report →

Coding

DeepSWE 1.0

62.0%

Coding

DeepSWE 1.1

53.0%

Long-horizon coding

SWE Marathon

29.0%

Full official benchmark table6 rows with source settings and peer charts

Benchmark	Area	Score	Setting / comparison
DeepSWE 1.0	Coding	62.0%	xAI reported software-engineering benchmark.
DeepSWE 1.1	Coding	53.0%	xAI reported software-engineering benchmark.
SWE Marathon	Long-horizon coding	29.0%	xAI reported long-horizon coding benchmark.
SWE Bench Pro output tokens	Efficiency	15,954 avg.	Average output tokens on SWE Bench Pro tasks.
Output-token efficiency	Efficiency	4.2x fewer than Opus 4.8	xAI comparison on SWE Bench Pro tasks.xAI compares Grok 4.5 against Opus 4.8 max on SWE Bench Pro tasks.
Serving speed	Latency	80 TPS	Reported generation speed.