PostTrainBench
PostTrainBench results reported for 16 models across 8 versions. Updated 2026-10-07.
Report an issueVersions
8 versions · Metric: PostTrainBench Weighted Aggregate
| Version | Released | Models | Results |
|---|---|---|---|
| PostTrainBench 1.1 — OpenCodeConfiguration: OpenCode harness | September 30, 2026 | 2 | 2 |
| PostTrainBench 1.2 — Claude CodeConfiguration: Claude Code harness | September 30, 2026 | 8 | 9 |
| PostTrainBench 1.2 — Codex CLIConfiguration: Codex harness | September 30, 2026 | 5 | 5 |
| PostTrainBench 1.2 — Cursor CLIConfiguration: Cursor CLI harness | September 30, 2026 | 1 | 1 |
| PostTrainBench 1.2 — OpenCodeConfiguration: OpenCode harness | September 30, 2026 | 1 | 1 |
| PostTrainBench 1.1 — Claude CodeConfiguration: Claude Code harness | July 28, 2026 | 5 | 6 |
| PostTrainBench 1.1 — Codex CLIConfiguration: Codex harness | July 28, 2026 | 3 | 3 |
| PostTrainBench 1.1 — Cursor CLIConfiguration: Cursor CLI harness | July 28, 2026 | 1 | 1 |
Recently reported results: PostTrainBench 1.1 — OpenCode, OpenCode harness
Most recent configuration: PostTrainBench 1.1 — OpenCode
| Model | Score | Source |
|---|---|---|
| Gemini 3.1 Pro | 22.0% | posttrainbench.com for Gemini 3.1 Pro on PostTrainBench 1.1 — OpenCode (opens in a new tab)Archived copy (opens in a new tab)PostTrainBench Weighted AggregateReported Benchmark leaderboardPostTrainBenchIndependentPostTrainBench TeamRecord for Gemini 3.1 Pro on PostTrainBench 1.1 — OpenCode |
| Gemini 4 Argon (highest thinking settings) | 45.3% | deepmind.google for Gemini 4 Argon (highest thinking settings) on PostTrainBench 1.1 — OpenCode (opens in a new tab)PostTrainBench Weighted AggregateReported Evaluation methodologyGoogle DeepMindSelf-reportedRecord for Gemini 4 Argon on PostTrainBench 1.1 — OpenCode |