GPT‑6‑Astra Solves 49 of 50 Physics Problems at $2.40 Each
Current overview (not historical)
GPT‑6‑Astra, a commercial AI agent developed by OpenAI, achieved a 98 % success rate on 49 of 50 theoretical physics research problems in the OpenAI Codex environment, with an average cost of under $3 per problem and a total benchmark cost below $116. The team tested the model without any human prompts; when network‑search capability was enabled, Astra performed best, whereas Gemini and GLM failed on black‑hole questions by reproducing a five‑dimensional formula incorrectly. This demonstrates a significant advance in AI problem‑solving for complex scientific domains while maintaining low operational costs.
Written with AI from the reporting · updated 10/15/2026
EventGraph Explorer
One time cutoff for reports, relations and evidence.
New at this step ·GPT-6-Astra Achieves 98% Accuracy in Theoretical Physics Benchmark
- Reports in replay
- 1
- First-party
- 0
- Sources
- 1
Reports first seen at this step · 1
- The Decoder · Third-partyGPT-6-Astra Achieves 98% Accuracy in Theoretical Physics BenchmarkOriginal source ↗
Discovery-time replay; the linked graph and evidence share this cutoff. The full overview and activity remain current.
Event developments
1 developments10/15/2026, 11:49:13 PM
GPT-6-Astra Achieves 98% Accuracy in Theoretical Physics BenchmarkThe Decoder
Filtered by first discovery; only representative reports already discovered by this cutoff are shown. Story grouping may have changed later, so this is not a complete historical snapshot.
Relations at this time
Loading time-scoped relations…
Bounded reconstruction, not an exact snapshot: entity names, relationship text and story grouping may include later revisions. Overview and full source archive below remain current.
Story route
A station-by-station route from the first public evidence onward; every station carries its own evidence and uncertainty.
- 1Oct 15, 00:00recorded Oct 15, 23:49single sourcerevised
GPT-6-Astra solves 49 of 50 physics research problems at $2.
Read original source ↗ The DecoderReplay this moment in the explorer
Uncertainty marks flag where to check the sources; they are not proven causation.
Full source archive (current)
Current public reports; outside the historical replay.
First-party sources · 0 reports
No verifiable reports in this group yet.
Independent coverage · 1 reports
- The Decoder · Oct 15, 23:24GPT-6-Astra Achieves 98% Accuracy in Theoretical Physics Benchmark
GPT-6-Astra solved 49 out of 50 theoretical physics problems in a benchmark test, with an average cost of under $3 per problem, demonstrating the potential of AI in scientific research.
Read original source ↗
Different reports do not necessarily disagree. Check original sources and publication dates.
Report timeline
Explore every side through dated source reports.
- The DecoderSelectedGPT-6-Astra Achieves 98% Accuracy in Theoretical Physics Benchmark
GPT-6-Astra solved 49 out of 50 theoretical physics problems in a benchmark test, with an average cost of under $3 per problem, demonstrating the potential of AI in scientific research.