LLM Perks

Gemini 2.5 Flash Thinking

Free right now Mid-size New
VisionReasoningCodingTool useLong context

Gemini 2.5 Flash Thinking is a language model from Google. It is currently free at 1 provider with up to 1M tokens of context. The model understands images, supports step-by-step reasoning, is suited to coding, supports tool calling.

Developer
Google
Context
1M tokens
Type
Text
License
Proprietary
Free since
October 6, 2026

Where to use Gemini 2.5 Flash Thinking for free

Each row is a separate free endpoint with its own limits.

ProviderContextLimitCard
UnoRouter
gemini-2.5-flash-thinking:free
1M No-card free models use shared capacity and per-model limits; availability can change. No Open

Quick start

curl https://api.unorouter.com/v1/chat/completions \
  -H "Authorization: Bearer $UNOROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-2.5-flash-thinking:free",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Example for UnoRouter. Any OpenAI-compatible client works: Cursor, Cline, OpenCode, Open WebUI.

Benchmarks

Intelligence Index
13.1
GPQA Diamond
79.0%
Humanity's Last Exam
12.1%
Terminal-Bench Hard
13.6%
τ²-Bench
31.6%
Long context (LCR)
65.3%

Source: Artificial Analysis · Gemini 2.5 Flash (Reasoning)

Similar free models

FAQ

Is Gemini 2.5 Flash Thinking free?

Yes. As of October 6, 2026, Gemini 2.5 Flash Thinking is free at UnoRouter. Each provider has its own limits — see the table above.

What is the context window of Gemini 2.5 Flash Thinking?

The largest context among free routes is 1,000,000 tokens. Some providers cap it lower.

How do I call Gemini 2.5 Flash Thinking via API?

Use any OpenAI-compatible client with base URL https://api.unorouter.com/v1, model ID gemini-2.5-flash-thinking:free, and a UnoRouter key.

How good is Gemini 2.5 Flash Thinking?

Artificial Analysis Intelligence Index: 13.1, GPQA Diamond: 79%. For comparison, the strongest free model in the catalog right now is GLM 5.3.

Is Gemini 2.5 Flash Thinking open-weight?

No, it is proprietary: access is only through provider APIs.