Google DeepMind · model passport
Gemini 1 5 Pro 002
gemini-1-5-pro-002- Cloud only
- Proprietary licence
- Released 2024-09-24(curated)
- Overall · LMArena
- 1,351#201 of 396
- Best percentile
- 53.6LiveBench
- Cohort standing
- 39.0median 38.1
What the data says
Benchmarked on 11 of 27 tracked axes. Every figure below is the model's own published result on that board.
Weakest axes
- 13 pct — Humanity's Last Exam (4.6%)
Uneven: 40.6 percentile points between its best and worst axis — 53.6 pct on LiveBench versus 13 pct on Humanity's Last Exam.
- LiveBench53.6 pct54.9%Verify →
- Instruction following50 pct1,339Verify →
- Overall49.5 pct1,351Verify →
- Hard prompts45.7 pct1,351Verify →
- GPQA Diamond41.6 pct57.2%Verify →
- Vision Arena39.3 pct1,179Verify →
- Coding37.9 pct1,356Verify →
- AIME (math)30.5 pct23.1%Verify →
- SimpleBench19.5 pct27.1%Verify →
- ARC-AGI-215.7 pct0.8%Verify →
- Humanity's Last Exam13 pct4.6%Verify →
- WebDev ArenaNot benchmarked
- Factuality ArenaNot benchmarked
- Search ArenaNot benchmarked
- Document ArenaNot benchmarked
- Agent ArenaNot benchmarked
- FrontierMathNot benchmarked
- SimpleQA VerifiedNot benchmarked
- BFCL v4 (function calling)Not benchmarked
- SWE-bench Verified (live board)Not benchmarked
- Terminal-Bench 4.0Not benchmarked
- MMLU-Pro (Artificial Analysis)Not benchmarked
- LiveCodeBench (Artificial Analysis)Not benchmarked
- IFBench (Artificial Analysis)Not benchmarked
- HHEM Hallucination (Vectara)Not benchmarked
- SWE-bench VerifiedNot benchmarked
- Terminal-BenchNot benchmarked
Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.
gemini-1.5-pro-002 stands at 39 (raw mean 36) — measured on 11 of 27 axes.
gemini-1.5-pro-002 stands at 39 versus a cohort median of 38.1 across 132 models released within 180 days — above the peer median.
On Humanity's Last Exam, gemini-1.5-pro-002 raised the running best score by 0.5 percentage points to 4.6% on 2024-09-24.
gemini-1.5-pro-002 scores in the 13th percentile on Humanity's Last Exam.
gemini-1.5-pro-002 ranks between #185 and #898 across 5 built-in weightings — best under "Conversation-first", worst under "All sources".
Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →