MDL
Gemini 3.1 Pro Preview
Verified against Artificial Analysis · Jul 27, 2026
Capability
97
percentile index
Price /M
$2 / $12
in / out
Context
1M
tokens
Avg score
71.0
7 benchmarks
AA Intelligence
46.5
index
Speed
132 tok/s
TTFT 31s
Released
Feb 19, 2026
About
Gemini 3.1 Pro Preview is a proprietary model in the Gemini family from Google. Benchmarked on 7 evals, averaging 71.0.
Benchmark scores · 7
avg 71.0 — higher is betterThe research behind this model
all papers mentioning it →OVO-S-Bench: A Hierarchical Benchmark for Streaming Spatial Intelligence in Multimodal LLMs▲ 31 on HF · Jun 2, 2026AutoMem: Automated Learning of Memory as a Cognitive Skill▲ 21 on HF · Jul 1, 2026S1-Omni: A Unified Multimodal Reasoning Model for Scientific Understanding, Prediction, and Generation▲ 16 on HF · Jul 17, 2026Unlocking the Visual Record of Materials Science: A Large-Scale Multimodal Dataset from Scientific Literature▲ 11 on HF · Jun 29, 2026Two-Level Meta-Rubrics for Evaluating Open-Ended Generation: GAMUT, a Benchmark for Factual Completeness▲ 9 on HF · Jul 21, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed132 tok/s
Latency (TTFT)31s
Price in / out$2 / $12 per 1M
Blended price$4.5 / 1M
Coding index68.8
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$12/M