Gemini 3.7 Flash
Aug 2026 · Proprietary · Current · supported evidence
Fastest serious reasoning model in the field. GPQA leader on several boards. Agentic depth still behind Claude and Grok.
Intelligence
56
AA Index
API price
$0.75 / $3.75
Input / output per 1M
Context
1M
330 tok/s
Capability
Intelligence Index56
Coding index76.1
Agentic index45.1
Arena Elo (offset)190
Use it when
- ▸Low-latency apps
- ▸Multimodal
- ▸Cheap GPQA-class Q&A
Skip if
- –Long-running agents are the job
Benchmarks
GPQA Diamond Graduate-level science questions designed so Google search is not enough. Still one of the cleanest knowledge/reasoning splits. | 94.8% |
|---|---|
Humanity’s Last Exam Expert-written questions across fields. Harder than MMLU; the current differentiator for “does this model actually know things.” | 47.9% |
MMLU-Pro Harder, less-saturated successor to MMLU. Classic MMLU is above 90% for every flagship and no longer ranks the field. | — |
SWE-bench Verified 500 human-validated GitHub issues. Score swings 5–15 points by harness — treat vendor numbers as an upper bound. | 80.8% |
SWE-bench Pro Harder, contamination-resistant coding eval. Currently the best public split between “can code” and “can maintain a repo.” | — |
Terminal-Bench 2.1 End-to-end tasks in a real terminal. Better proxy for coding agents than HumanEval, which is fully saturated. | — |
ARC-AGI-2 Abstract visual puzzles. Rewards generalization over memorization. GPT-5.6 Sol currently leads the published set. | 84.6% |
AIME 2025 American Invitational Mathematics Examination. Contest math; reasoning-mode models dominate. | 97.2% |
MMMU College-level multimodal understanding across diagrams, charts, and exam figures. | — |
Modalities
text · vision · audio · video · tools
Reasoning mode
hybrid
Hybrid and reasoning models spend tokens thinking. That raises GPQA and agents, and also raises latency and bill.