Gemini 3.5 Flash Lite Reaches General Availability
Google released Gemini 3.5 Flash Lite as a stable, production-ready model on July 21, 2026. This brief outlines its confirmed specs and evaluation steps.
Official Release Details
Google officially designated Gemini 3.5 Flash Lite as generally available on July 21, 2026. This date marks the release of a stable, production-ready version of the model. The announcement occurred simultaneously with the general availability of Gemini 3.6 Flash. According to the official changelog, this specific iteration is designed as a subagent option intended for high-volume automation scenarios.
- Release Date: July 21, 2026
- Status: Generally Available (GA)
- Provider: Google
- Model ID: gemini-3.5-flash-lite
Confirmed Technical Specifications
The source documentation confirms that Gemini 3.5 Flash Lite accepts both text and image modalities for input. Its output modality is restricted to text only. The model features a context window of 1,048,576 tokens and a maximum output limit of 65,536 tokens. The record indicates that open weights are not confirmed for this release. No pricing data is currently listed in the provided facts. The description explicitly characterizes the model as offering low latency and high cost-effectiveness for automated processes.
- Input Modalities: Text, Image
- Output Modality: Text
- Context Window: 1,048,576 tokens
- Max Output Tokens: 65,536
- Open Weights: Not Confirmed
Evaluation Procedure for Engineers
Engineers should validate the model’s performance against their specific automation requirements using a structured testing protocol. First, test the model by submitting representative batches of text and image inputs to verify processing stability. Measure the response time and token usage during these runs to assess whether the low-latency claim holds under your specific load conditions. Compare the results against your existing baseline systems or previous model versions to determine if the cost-efficiency profile meets project constraints. Finally, validate the output quality by reviewing the generated text for accuracy and relevance within the 65,536-token output limit. Note that the source does not provide benchmark scores, so all performance conclusions must be derived from your own empirical measurements.
Artificial Analysis benchmark review
Benchmark review updated on . These scores apply to the linked artifact and recorded methodology; they do not establish the performance of a different runtime or quantized model.
| Measurement | Recorded result | Measured |
|---|---|---|
| Artificial Analysis Coding Index (v4.1) ↗ | 49.3 index_points | 2026-08-09 |
| Artificial Analysis Coding Index (v4.2) ↗ | 49.3 index_points | 2026-09-05 |
| Artificial Analysis Coding Index (v4.3) ↗ | 49.3 index_points | 2026-09-20 |
| Artificial Analysis Intelligence Index (v4.1) ↗ | 37.4 index_points | 2026-08-09 |
| Artificial Analysis Intelligence Index (v4.2) ↗ | 27.6 index_points | 2026-09-05 |
| Artificial Analysis Intelligence Index (v4.3) ↗ | 22.2 index_points | 2026-09-20 |
Review the full benchmark provenance and source methodology →
Primary sources
Frequently asked questions
What is the exact date of the Gemini 3.5 Flash Lite announcement?
The official announcement date recorded in the source is July 21, 2026. This date refers to the historical event of the general availability release, distinct from the current publication date of this article.
Does the source confirm local installation capabilities?
No. The source states that open weights are not confirmed. It does not establish that local installation is impossible, nor does it mandate API access. The documentation only records the lack of confirmed open weights without specifying hosting arrangements or deployment architecture.