THE FORGE RANKINGS·2026-08-30

GPT-5.2 (2025-12-11)

OpenAI
19
RANK of 37 ranked
FERROX INDEX65.5 weighted mean of category positions
GRADEC below the field median
EVIDENCE9 / 32 benchmarks measured
QUALITYFULL every category rests on independent benchmarks

Not separable from Kimi K2.6 at the measured 1.78 point band. The order is printed; the gap is not claimed.

No measurement in Preference. That weight was redistributed, which is the same as assuming this model would have scored its own average there. It is an assumption, not a measurement.

Compared to what

Each rail is one classification. Every ranked model measured on it is a tick; this model is the orange marker. A hatched rail means this model has no measurement in that classification and its weight was redistributed.

Agents

15of 37 measured
43.0
13.1median 38.963.0

Coding

9of 37 measured
54.7
22.4median 50.070.7

Reasoning

24of 37 measured
56.1
18.1median 62.893.0

Preference

no qualifying measurement
53.1median 71.882.5

Every measurement, and the field behind it

One card per benchmark this model has been measured on. The rail shows where it sits against every other ranked model measured on the same benchmark. Where the evaluator keys on a harness, the harness is named, because the same model scores differently under a different one.

SWE-bench Verified (Epoch's own run)

10/21
73.8% best 83.5% · Claude Opus 4.7
Epoch AI CC-BY-4.0 high

Humanity's Last Exam

5/17
27.8% best 46.4% · Gemini 3.1 Pro Pr…
Humanity’s Last Exam (CAIS / Scale AI) CC-BY-4.0

Not measured

23/32
  • LMArena Agent
  • LMArena Text (style-controlled)
  • DeepSWE
  • FrontierCode
  • LMArena WebDev
  • METR time horizons
  • Terminal-Bench 4.0
  • Aider Polyglot
  • OSWorld
  • Cybench
  • BrowseComp
  • IMOAnswerBench
  • SWE-bench Multilingual
  • AIME
  • CyberGym
  • HMMT
  • MCP-Atlas
  • SWE-bench Pro
  • Tau2-Bench Airline
  • Tau2-Bench Banking
  • Tau2-Bench Retail
  • Tau2-Bench Telecom
  • Tool-Decathlon
A missing benchmark is not a zero and is never scored as one.

Check it yourself

Every number on this page, with who measured it, the interval they published, the harness it was run under, and where to go and read it. Nothing here was measured by Ferrox Labs.

BenchmarkClassPublishedInterval (normalised) Measured byLicenceHarness / effort
APEX Agents agents 23.0% 16.7 to 29.3 APEX CC-BY-4.0 / high source
Terminal-Bench 2 agents 62.9% 57.0 to 68.8 Terminal-Bench v2 Leaderboard CC-BY-4.0 Codex CLI / medium source
ALE-Bench coding 1250 not published ALE-Bench (Sakana AI with AtCoder) CC-BY-4.0 / medium source
SWE-bench Verified (Epoch's own run) coding 73.8% 69.8 to 77.7 Epoch AI CC-BY-4.0 / high source
ARC-AGI-2 reasoning 43.3% not published ARC Prize CC-BY-4.0 / high source
FrontierMath reasoning 36.9% 31.3 to 42.5 Epoch AI CC-BY-4.0 / medium source
GPQA Diamond (Epoch's own run) reasoning 87.9% 84.1 to 91.6 Epoch AI CC-BY-4.0 / medium source
Humanity's Last Exam reasoning 27.8% 24.4 to 31.3 Humanity’s Last Exam (CAIS / Scale AI) CC-BY-4.0 source
OTIS Mock AIME 2024-2025 reasoning 93.9% 87.9 to 99.9 Epoch AI CC-BY-4.0 / medium source

Configurations rolled up: 8. Rule: median observed configuration per benchmark (lower median, always a real measurement). Harnesses seen: codex-cli, droid.

Snapshot 2026-08-30-62e4e85f043b, manifest dc46131315046f1e. Grades are positional: position in the measured field, as a percentile of rank among ranked entries, n=37.