Google
Gemini 3.5 Flash
Google's speed/quality sweet spot — near-flagship scores at Flash pricing.
API onlytextimagevideofileaudioREL 2026-05-19
Compare this model →Verified 2026-06-12
Overview
Gemini 3.5 Flash punches far above its 'Flash' branding: 78.8% on SWE-bench Verified and ~55 on the Artificial Analysis Intelligence Index — within reach of flagships costing 3–5× more. Full multimodality (text, image, video, audio, file) with a 1M context at $1.50/$9 per 1M tokens.
Benchmark scores
Methodology →Intelligence55
Coding78.8
Reasoningn/a
Mathn/a
Sources and verification
- Coding: SWE-bench Verified · checked 2026-06-09
- Intelligence: Artificial Analysis Intelligence Index · checked 2026-06-12
- Catalogue and pricing: OpenRouter model record · checked 2026-06-12
Strengths and weaknesses
Strengths
- Best price/performance of any frontier model
- Full multimodality incl. video + audio
- Fast
Trade-offs
- 65K max output — smaller than rivals
- Closed weights
Reach for this model when
The default pick for most workloads — coding, multimodal analysis, agents on a budget.
Run something like it locally
Open-weights peers: Qwen3.7-Plus or MiniMax M3 via vLLM, though neither matches the multimodal range.
