name = "Grok 4.5"
description = "xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk"
family = "grok"
release_date = "2026-07-08"
last_updated = "2026-07-08"
attachment = true
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = false

[limit]
context = 500_000
output = 500_000

[modalities]
input = ["text", "image"]
output = ["text"]

[[benchmarks]]
name = "SWE-Bench Pro"
score = 64.7
metric = "resolve rate"
source = "https://x.ai/news/grok-4-5"
date = "2026-07-08"

[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 78
metric = "resolve rate"
source = "https://x.ai/news/grok-4-5"
date = "2026-07-08"

[[benchmarks]]
name = "Terminal-Bench"
score = 83.3
metric = "success rate"
version = "2.1"
source = "https://x.ai/news/grok-4-5"
date = "2026-07-08"

[[benchmarks]]
name = "DeepSWE"
score = 62.0
metric = "resolve rate"
version = "1.0"
source = "https://x.ai/news/grok-4-5"
date = "2026-07-08"

[[benchmarks]]
name = "DeepSWE"
score = 53
metric = "resolve rate"
harness = "mini-swe-agent"
version = "1.1"
source = "https://x.ai/news/grok-4-5"
date = "2026-07-08"

[[benchmarks]]
name = "SWE Marathon"
score = 29.0
metric = "pass@1"
source = "https://x.ai/news/grok-4-5"
date = "2026-07-08"