name = "Google Gemini 2.5 Flash Lite"
description = "Low-latency Gemini model for high-volume multimodal and agent workloads"
# Native exception (accessed 2026-06-25): `thinkingBudget` 0 disables and -1 is
# dynamic; enabled manual budgets are 512..24576 tokens.
# https://ai.google.dev/gemini-api/docs/thinking
family = "gemini-flash-lite"
release_date = "2025-07-22"
last_updated = "2025-07-22"
attachment = false
reasoning = true
reasoning_options = [{ type = "toggle" }, { type = "budget_tokens", min = 512, max = 24_576 }]
temperature = true
tool_call = true
knowledge = "2025-07"
open_weights = false

[cost]
input = 0.09999999999999999
output = 0.39999999999999997
cache_read = 0.024999999999999998
cache_write = 0.09999999999999999

[limit]
context = 1048576
output = 65535

[modalities]
input = ["text", "image"]
output = ["text"]