The Open-Source AI Stack

RSS

Models · gemini

Gemini 1.5 Pro

Proprietary Google DeepMind · 2024-02-15 · Proprietary

Google's first long-context Gemini checkpoint, introduced with a 128K standard window and a 1M token preview tier. Google described the design as a mixture-of-experts that activates a subset of expert networks per input, and demonstrated 99% recall on needle-in-a-haystack across 1M tokens at launch. The context window was later extended to 2M tokens in private preview, announced May 14 2024.

Cost

$0.00 / Mtok input

$0.00 / Mtok output

· as of 2026-05-21

Speed

0 tok/sec output

0 ms TTFT

· as of 2026-05-21

Architecture

Schema-generated from data/models.yaml. Every label is auditable against the model's sources.

Specs

Architecture: unknown
Total params: not disclosed
Active params: not disclosed
Context window: 2.1M tokens
Attention: unknown
Position encoding: unknown
Post-training: rlhf
OSI-approved: no
Data released: no
Training code: not released

Benchmarks

Each score carries the date it was published; we never infer or interpolate missing scores.

General reasoning

MMLU-Pro	75.0	as of 2026-05-21	source ↗
GPQA-Diamond	58.9	as of 2026-05-21	source ↗

Code

LiveCodeBench

31.6

as of 2026-05-21

Math

MATH	87.6	as of 2026-05-21	source ↗
AIME 2024	23.0	as of 2026-05-21	source ↗

Available quantizations

None. The weights are not distributed, so there are no public quantizations.

Notable innovations

· Mixture-of-experts Gemini architecture
· 1M token context at launch, later 2M
· Strong in-context language learning (Kalamang grammar)

Sources