Note: Overall leaderboard rankings may not reflect true model quality — individual benchmarks give a clearer picture. ARC-Challenge MMLU GPQA GSM8K Artificial Analysis Intelligence Index v4.1
← Back to leaderboard

Google: gemini-3-flash

6 benchmarks
CompanyGoogle
Country of originUSA

Leaderboard position

#140
Overall rank (of 1122 models)
2076
Elo rating
6
Benchmarks
Top 12.5%
Percentile
In Google (of 73 models)
Above: OpenAI: gpt-5.2-codex #139 Below: OpenAI: gpt-5.1 #141

Benchmark highlights

Nearest competitor from another company

The model from another company with the most similar benchmark scores.

appeared as nearest in 1 of 6 benchmarks

Benchmarks

Predicted scores

Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.

Benchmark Predicted Confidence Based on
Humanity's Last Exam (no tools) ~33.43 60% Moonshot AI: Kimi K2.6, DeepSeek: deepseek-v3.2
SciCode ~46.03 60% Moonshot AI: Kimi K2.6
Humanity's Last Exam (with tools) ~45.04 60% DeepSeek: deepseek-v3.2
LiveCodeBench v6 ~83.89 60% Moonshot AI: Kimi K2.6
MMMU-Pro ~75.92 60% Moonshot AI: Kimi K2.6
BrowseComp ~71.25 60% Moonshot AI: Kimi K2.6
GPQA ~84.17 60% Moonshot AI: Kimi K2.6, DeepSeek: deepseek-v3.2
ScreenSpot-Pro ~79 60% Qwen: qwen3.5-397b-a17b
Terminal-Bench v2.0 ~55.69 59% Moonshot AI: Kimi K2.6, Z.ai: GLM 5, MiniMax: minimax-m2.7
AIME 2026 ~97.08 58% Moonshot AI: Kimi K2.6, Z.ai: GLM 5, Qwen: qwen3.5-397b-a17b
HMMT Feb 2026 ~88.77 58% Moonshot AI: Kimi K2.6, Z.ai: GLM 5, Qwen: qwen3.5-397b-a17b
MMLU-Pro ~90.53 58% Z.ai: GLM 5, Qwen: qwen3.5-397b-a17b, DeepSeek: deepseek-v3.2
SWE-bench Verified ~77.53 58% Moonshot AI: Kimi K2.6, Z.ai: GLM 5, MiniMax: minimax-m2.7
SWE-bench Pro ~53.43 58% Moonshot AI: Kimi K2.6, MiniMax: minimax-m2.7, DeepSeek: deepseek-v3.2
Artificial Analysis Intelligence Index v4.0 (Maximum Reasoning) ~43.69 58% Moonshot AI: Kimi K2.6, Z.ai: GLM 5, OpenAI: gpt-5.2
GPQA Diamond ~84.86 57% Z.ai: GLM 5
IFBench ~73.54 57% Z.ai: GLM 5
Terminal-Bench v2.1 ~63.55 57% MiniMax: minimax-m2.7
Artificial Analysis Intelligence Index v4.1 (Maximum Reasoning) ~36.57 57% Moonshot AI: Kimi K2.6, Z.ai: GLM 5, OpenAI: gpt-5.2
AIME 2025 ~100 56% Z.ai: GLM 5, Google: Gemini 3 Pro Preview, DeepSeek: deepseek-v3.2
Claw-Eval ~64.53 56% Moonshot AI: Kimi K2.6, MiniMax: minimax-m2.7, Qwen: qwen3.5-397b-a17b
MMLU-Redux ~95.6 56% DeepSeek: deepseek-v3.2
Text Arena (Snapshot from 20 june 2026) ~1453.21 55% Google: Gemini 3 Pro Preview
LLMArena (llmarena.ru) (Snapshot from 19 june 2026) ~1125.03 55% Google: Gemini 3 Pro Preview, DeepSeek: deepseek-v3.2
SuperGPQA ~70.03 54% OpenAI: gpt-5.2, Google: Gemini 3 Pro Preview, Qwen: qwen3.5-397b-a17b