Google DeepMind · closed · 2026-05
Gemini 3.5 Flash
not ranked n/a
the takeGoogle's newest Flash, tuned for agents and, by their own tables, beating last year's Pro. Off the board because they withheld the standard academic evals. The vendor charts look great; I want the neutral ones.
Not ranked: Google DeepMind doesn't publish enough comparable public benchmarks to place it fairly. How ranking works.
MMLU-Pro
—
- Context
- 1.048576M
- Input
- $1.5/M
- Output
- $9/M
- Speed
- 167.4 tok/s
- Modality
- text, image, audio, video
- Frontier agentic/tool-use for its tier
- Strong multimodal reasoning
- Cheaper and faster than Pro
- Withheld standard academic benchmarks
- Still trails Pro on hardest reasoning
- ~3x the price of Gemini 3 Flash