GPT-6 AstravsGPT-5.4
Across 8 shared benchmarks, GPT-6 Astra leads overall: GPT-6 Astra wins 8, GPT-5.4 wins 0, with 0 ties and an average score difference of +23.91.
GPT-6 Astra
OpenAI · 2026-09-03 · Reasoning model
GPT-5.4
OpenAI · 2026-03-05 · Multimodal model
GPT-6 Astra8 wins(100%)(0%)0 winsGPT-5.4
Benchmark scores
Grouped by capability, sorted by largest gap within each. 8 shared benchmarks.
General Knowledge
GPT-6 Astra 4/4| Benchmark | GPT-6 Astra | GPT-5.4 | Diff |
|---|---|---|---|
| ARC-AGI-3 | 62.701 / 16Max (No Tools) | 014 / 16Thinking High (No Tools) | +62.70 |
| ARC-AGI-2 | 951 / 85Max (No Tools) | 77.1023 / 85Normal (No Tools) | +17.90 |
| HLE | 57.2012 / 190Max (With Tools) | 52.1028 / 190Extra-High (With Tools) | +5.10 |
| ARC-AGI-1 | 98.501 / 91Max (No Tools) | 93.7021 / 91Normal (No Tools) | +4.80 |
AI Agent - Information Search
GPT-6 Astra 1/1| Benchmark | GPT-6 Astra | GPT-5.4 | Diff |
|---|---|---|---|
| BrowseComp | 91.501 / 56Max (With Tools) | 82.7016 / 56Extra-High (With Tools) | +8.80 |
Coding and Software Engineer
GPT-6 Astra 1/1| Benchmark | GPT-6 Astra | GPT-5.4 | Diff |
|---|---|---|---|
| DeepSWE | 74.102 / 35Max (With Tools) | 5227 / 35Extra-High (With Tools) | +22.10 |
General Evaluation
GPT-6 Astra 1/1| Benchmark | GPT-6 Astra | GPT-5.4 | Diff |
|---|---|---|---|
| GPQA Diamond | 961 / 271Max (No Tools) | 74.75174 / 271Normal (No Tools) | +21.25 |
Math and Reasoning
GPT-6 Astra 1/1| Benchmark | GPT-6 Astra | GPT-5.4 | Diff |
|---|---|---|---|
| FrontierMath Tier 4 v2 | 97.601 / 41Max (No Tools) | 4911 / 41Extra-High (No Tools) | +48.60 |
Specs
| Field | GPT-6 Astra | GPT-5.4 |
|---|---|---|
| Publisher | OpenAI | OpenAI |
| Release date | 2026-09-03 | 2026-03-05 |
| Model type | Reasoning model | Multimodal model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1.05M | 1M |
| Max output | 128K | 125K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-6 Astra | GPT-5.4 |
|---|---|---|
| Text input | $10 / 1M tokens | $2.5 / 1M tokens |
| Text output | $50 / 1M tokens | $15 / 1M tokens |
| Cache read | $1 / 1M tokens | $0.25 / 1M tokens |
| Cache write | $12.5 / 1M tokens | Not public |
Summary
- GPT-6 Astraleads in:General Knowledge (4/4), AI Agent - Information Search (1/1), Coding and Software Engineer (1/1), General Evaluation (1/1), Math and Reasoning (1/1)
On average across the 8 shared benchmarks, GPT-6 Astra scores 23.91 higher.
Largest single-benchmark gap: ARC-AGI-3 — GPT-6 Astra 62.70 vs GPT-5.4 0 (+62.70).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.