Flash 3.5
#60 in Frontier-Spraakmodellegoogle · v3.5 · siet 2026-05-19 · 6× · tolest 30. Juni 2026
24
Momentum
Gemini 3.5 Flash is a proprietary frontier language model by Google DeepMind, released to general availability on May 19, 2026 at Google I/O 2026 as the first model in the Gemini 3.5 family. It is the strongest agentic and coding-focused Flash model in the Gemini lineup to date. The model natively supports multimodal input (text, image, audio, video, PDF) with a 1-million-token context window, and is accessible via the Gemini API, Google AI Studio, Vertex AI, the Gemini app, and Google Search AI Mode. API pricing was tripled compared to its predecessor Gemini 3 Flash, but sits approximately 40% below the cost of Gemini 3.1 Pro.
Momentum-Verloop
04.04.03.07.
Features
| Coding Benchmark (SWE-bench Pro %) | 55.1% on SWE-Bench Pro (for comparison: Claude Opus 4.7 leads with 64.3%) |
| Context Window (Tokens) | 1,048,576 input tokens (~1M); max. 65,536 output tokens |
| Multimodal (Image/Video) | Yes – native multimodality: text, image, audio, video, PDF as input; text-only output |
| Price per 1M Input Tokens | $1.50 / 1M input tokens (standard, global region); $0.15 / 1M for cached tokens; $1.65 / 1M in non-global regions |
| Price Tier (Monthly) | Free in Gemini App and AI Mode in Google Search; Gemini AI Plus / AI Pro: $20/month (includes access to 3.5 Flash); API: pay-per-token (no mandatory subscription) |