wehub-resource-sync
106 lines
2.2 KiB
TOML
106 lines
2.2 KiB
TOML
name = "GPT-5.4 Pro"
|
|
description = "More exact GPT-5.4 tier for demanding professional reasoning and agent tasks"
|
|
family = "gpt-pro"
|
|
release_date = "2026-03-05"
|
|
last_updated = "2026-03-05"
|
|
attachment = true
|
|
reasoning = true
|
|
temperature = false
|
|
tool_call = true
|
|
structured_output = false
|
|
knowledge = "2025-08-31"
|
|
open_weights = false
|
|
|
|
[limit]
|
|
context = 1_050_000
|
|
input = 922_000
|
|
output = 128_000
|
|
|
|
[modalities]
|
|
input = ["text", "image"]
|
|
output = ["text"]
|
|
|
|
[[benchmarks]]
|
|
name = "GPQA Diamond"
|
|
score = 94.4
|
|
metric = "accuracy"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "Humanity's Last Exam"
|
|
score = 42.7
|
|
metric = "accuracy"
|
|
variant = "no tools"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "Humanity's Last Exam"
|
|
score = 58.7
|
|
metric = "accuracy"
|
|
variant = "with tools"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "BrowseComp"
|
|
score = 89.3
|
|
metric = "accuracy"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "GDPval"
|
|
score = 82.0
|
|
metric = "wins or ties"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "FrontierMath"
|
|
score = 50.0
|
|
metric = "accuracy"
|
|
dataset = "Tier 1-3"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "FrontierMath"
|
|
score = 38.0
|
|
metric = "accuracy"
|
|
dataset = "Tier 4"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "ARC-AGI-1"
|
|
score = 94.5
|
|
metric = "accuracy"
|
|
variant = "Verified"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "ARC-AGI-2"
|
|
score = 83.3
|
|
metric = "accuracy"
|
|
variant = "Verified"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "FinanceAgent"
|
|
score = 61.5
|
|
metric = "accuracy"
|
|
version = "1.1"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|
|
|
|
[[benchmarks]]
|
|
name = "GeneBench"
|
|
score = 25.6
|
|
metric = "accuracy"
|
|
source = "https://openai.com/index/introducing-gpt-5-5/"
|
|
date = "2026-04-23"
|