GRB internal research engineering
Benchmark by Google DeepMind
August 2026 report snapshot
27%
Gemini 3.7 Flash / GRB scaffold
frontier lab
GRB, a qualitative RE-Bench assessment and AlphaEvolve. Different internal metrics cannot rank labs.
Benchmark by Google DeepMind
August 2026 report snapshot
27%
Gemini 3.7 Flash / GRB scaffold
Benchmark by Google DeepMind
August 2026 report snapshot
around 16%
Gemini 3.1 Pro / preliminary GRB runs
Benchmark by OpenAI
Original paper v1
3.2 ± 0.2%
Gemini 2.0 Flash / BasicAgent
Benchmark by METR
Google five-task subset, November 2025
Qualitative result
Gemini 3 Pro / METR Modular adaptation
Measure by Google DeepMind
May 2025 report
1%
AlphaEvolve / Gemini ensemble
Measure by Google DeepMind
May 2025 report
23%
AlphaEvolve / Gemini ensemble
Benchmark by METR
Google model-card assessment, February 2026
1.27 normalized score
Gemini 3.1 Pro (Deep Think)
Benchmark by METR
Google model-card assessment, February 2026
1.04 normalized score
Gemini 3 Pro (February 2026 comparison)
Benchmark by PostTrainBench team
PostTrainBench v1.1 · September 2026 results
21.99%
Gemini 3.1 Pro / OpenCode
Benchmark by PostTrainBench team
PostTrainBench v1.2 · September 30 release
18.11%
Gemini 3.1 Pro / OpenCode
Benchmark by PostTrainBench team
PostTrainBench v1.1 / Google OpenCode evaluation, October 2026
45.3%
Gemini 4 Argon / Google OpenCode evaluation
risk report · 2026-08 · available
risk report · 2025-11 · available
organizational research article · 2025-05-14 · available
system card · 2026-02-19 · available
other primary · 2026-10 · available
original paper · 2026-09-25 · available