Models Frontier Benchmarks Prices Research Graph Changes

Loading timeline…

AI Atlas

The temporal knowledge graph of the AI ecosystem. Every number links back to its source, tier and observation time; history is never discarded; conflicts are shown, never averaged.

www.ai-atlas.co/api/v1

Explore

Models
Companies
Research
Providers & Pricing
Benchmarks
Hardware
Frameworks
Datasets
Tools
Open models
Families

Intelligence

Frontier
Compare
Prices
Calculator
Run locally
Find a model
Knowledge graph
Pulse

Temporal

Changes
Timeline
Time machine
Diff two dates
Watchlist

About

About
Methodology
Sources
API / Developers
AIAtlasBot

Built by Simon-Pierre Boucher · contact@spboucher.ai · Hosted on MacLustr (www.maclustr.io)

Home
Models
Changes

Model Timeline

Timeline of gpt-5-4-nano-non-reasoning

Every recorded change for this model by OpenAI, keyed on when it occurred. Back to the entity →

24 events

Filters

Year

All years 2026

Category (lane)Event typeFilters the loaded events (the API has no type filter on /timeline).OrganizationFamilyOrganization and family scope the timeline to that entity (one at a time).Min importanceInclude historical backfillBackfill = history imported when a source is first crawled (release dates, past prices). Off by default so the timeline shows what AI Atlas observed happening; on, it shows the reconstructed history.

Reset

Lanes · dots sized by importance · drag to select a rangeoccurred vs observed vs recordedOccurred — coalesce(effective_at, observed_at) — when the change happened (effective date when a source states it)Observed — when AI Atlas first saw the changeRecorded — when the row was writtenBackfill — true for history imported when a source is first crawled (never shown as 'today' in feeds)

gpt-5.4-nano scores 33.33% on Terminal-Bench
gpt-5.4-nano scores 52.63% on τ²-bench
gpt-5.4-nano scores 59.54% on MMMU-Pro
gpt-5.4-nano scores 64.42% on IFBench
gpt-5.4-nano scores 15.94% on Humanity's Last Exam
gpt-5.4-nano scores 76.06% on GPQA Diamond
gpt-5.4-nano scores 20.01 on Artificial Analysis Intelligence Index
gpt-5.4-nano scores 24.24% on Terminal-Bench
gpt-5.4-nano scores 34.8% on τ²-bench
gpt-5.4-nano scores 43.76% on MMMU-Pro
gpt-5.4-nano scores 32.72% on IFBench
gpt-5.4-nano scores 4.08% on Humanity's Last Exam
gpt-5.4-nano scores 55.76% on GPQA Diamond
gpt-5.4-nano scores 11.69 on Artificial Analysis Intelligence Index
gpt-5.4-nano scores 42.42% on Terminal-Bench
gpt-5.4-nano scores 60.67% on Terminal-Bench
gpt-5.4-nano scores 0.51% on Terminal-Bench
gpt-5.4-nano scores 76.02% on τ²-bench
gpt-5.4-nano scores 65.38% on MMMU-Pro
gpt-5.4-nano scores 75.92% on IFBench
gpt-5.4-nano scores 47.22% on SciCode
gpt-5.4-nano scores 28.27% on Humanity's Last Exam
gpt-5.4-nano scores 81.72% on GPQA Diamond
gpt-5.4-nano scores 21.24 on Artificial Analysis Intelligence Index

Benchmarks12 Sept 2026 → 12 Sept 2026

24 events loaded · every dot is one event; hover for its summary

Sept 202624

Sept 2026 24digest →

12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 33.33% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 52.63% on τ²-bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 59.54% on MMMU-Pro
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 64.42% on IFBench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 15.94% on Humanity's Last Exam
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 76.06% on GPQA Diamond
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 20.01 on Artificial Analysis Intelligence Index
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 24.24% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 34.8% on τ²-bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 43.76% on MMMU-Pro
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 32.72% on IFBench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 4.08% on Humanity's Last Exam
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 55.76% on GPQA Diamond
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 11.69 on Artificial Analysis Intelligence Index
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 42.42% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 60.67% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 0.51% on Terminal-Bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 76.02% on τ²-bench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 65.38% on MMMU-Pro
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 75.92% on IFBench
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 47.22% on SciCode
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 28.27% on Humanity's Last Exam
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 81.72% on GPQA Diamond
artificial_analysis
artificialanalysis.ai
12 Sept 2026
Benchmark resultModelgpt-5.4-nanoOpenAI
gpt-5.4-nano scores 21.24 on Artificial Analysis Intelligence Index
artificial_analysis
artificialanalysis.ai

Legend

Lanes

Models0
Research0
Prices0
Benchmarks24
Providers0
Hardware0
Frameworks0
Companies0

Semantics

occurred_at: coalesce(effective_at, observed_at) — when the change happened (effective date when a source states it)
observed_at: when AI Atlas first saw the change
recorded_at: when the row was written
is_backfill: true for history imported when a source is first crawled (never shown as 'today' in feeds)
group_key: one release / announcement seen through several documents shares a group_key

Times are UTC. Per-entity history: open any entity's History tab. Compare two dates in Diff; see the atlas as of a date in the Time machine.

Showing up to 1,000 events. Dates are occurred_at (effective date when a source states one, else observation) — hover a date for the observation time. Historical backfill is excluded; the checkbox in the rail includes it. Cursor-paged history: changes feed.

Earliest loaded event: 2026-09-12T07:08:16.635021Z