Gemini 3 Flash in GitHub Copilot

Google - Lightweight- preview

In GitHub Copilot, Gemini 3 Flash costs $0.5 per 1M input tokens and $3 per 1M output tokens, billed as AI credits at $0.01 each. A typical daily-coding request works out to about 0.7 credits (about $0.01) with auto model selection on. Rates verified against GitHub's rate sheet on 2026-07-14.

Token rates

Input / 1M
$0.50
Cached input / 1M
$0.05
Output / 1M
$3.00
Cost per credit
$0.01

Credits per request

Defined workload baskets, not measured tasks, with auto model selection on.

WorkloadCredits (auto on)Credits (auto off)Cost
Short chat turn0.40.4$0.00
Daily coding request0.70.8$0.01
Agent session step2.42.6$0.02

At about 0.7 credits per daily-coding request, Gemini 3 Flash ranks #4 of 22 current models by cost, about 2.4x the cheapest (GPT-5.4 nano).

Requests per plan allowance

Daily-coding requests the included monthly credits buy, before any overage.

PlanCredits / monthRequests
Copilot Pro1,5002,136
Copilot Pro+7,0009,971
Copilot Max20,00028,490
Copilot Business1,9002,706
Copilot Enterprise3,9005,555

Copilot rate vs direct API

GitHub Copilot bills Gemini 3 Flash at $0.50 in / $3.00 out per 1M. That input rate is the same as the direct Google gemini-3-flash-preview API price of $0.50 in / $3.00 out (validated 2026-06-15).

Estimate your monthly burn with Gemini 3 Flash

Cheaper: Raptor mini - Pricier: GPT-5.4 mini

All models - GitHub AI Credits