NEWMEN
← back to all models

Google: Gemini 2.5 Pro

google/gemini-2.5-pro

Googleclosed-weight1.0M context4 providersIntelligence 84.6

Cheapest provider

$1.25 / 1M

Google

Fastest provider (p95)

148 tok/s

Google AI Studio

Intelligence (composite)

84.6

MMLU-Pro · HumanEval · math · GPQA

Per-provider performance

Latency / throughput / uptime measured across providers over the last 30 minutes of live traffic. Atlas’s router weighs these per call (with the eval-gate signal) when picking a variant for Standard and Batch tiers.

ProviderQuantInput $/1MOutput $/1MLatency p50 / p95Throughput p50 / p95Uptime 30mSuccess
Googlefull· unknown$1.2500$10.00009932ms / 30083ms91 / 128 tok/s96.73%33.5%
Googlefull· unknown$1.2500$10.00009932ms / 30083ms91 / 128 tok/s61.36%33.5%
Googlefull· unknown$1.2500$10.00009932ms / 30083ms91 / 128 tok/s29.41%33.5%
Google AI Studiofull· unknown$1.2500$10.00003104ms / 18651ms105 / 148 tok/s85.40%90.5%

“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).

Intelligence breakdown

Composite score is a weighted average of public benchmarks (30% MMLU-Pro, 25% code pass@1, 25% math, 20% GPQA). Numbers come from model cards and the Artificial Analysis intelligence harness; missing components are renormalised over what’s present.

MMLU-Pro

86.0

broad reasoning

Code

80.0

pass@1 (HumanEval / LiveCodeBench)

AIME 2025

88.0

math accuracy

GPQA Diamond

84.0

hard reasoning

Source: Artificial Analysis (AA-aligned) — Gemini 2.5 Pro (MMLU-Pro 86, GPQA-D 84, LiveCodeBench 80, AIME25 88); github.com/MiniMax-AI/MiniMax-M2

How Atlas routes Google: Gemini 2.5 Pro

  • Realtime tier — direct passthrough at the upstream’s native precision. Best for hard latency / quality guarantees.
  • Standard tier — Atlas picks the cheapest provider variant whose quantization has stayed green on your operation’s eval gates. For Google: Gemini 2.5 Pro that’s currently Google at $1.25/1M.
  • Batch tier — async, biggest discount. Roadmapped to use provider batch APIs (OpenAI / Anthropic) where available and queued spot capacity for open-weight workloads.