MDL
Gemini 2.5 Pro
Verified against Artificial Analysis · Jul 13, 2026
Capability
82
percentile index
Price /M
$1.25 / $10
in / out
Context
1M
tokens
Avg score
64.7
12 benchmarks
AA Intelligence
25.8
index
Speed
140 tok/s
TTFT 18s
Released
Jun 5, 2025
About
Gemini 2.5 Pro is a proprietary model in the Gemini 2.5 family from Google. Benchmarked on 12 evals, averaging 64.7.
proprietarytextvision
Benchmark scores · 12
avg 64.7 — higher is betterThe research behind this model
all papers mentioning it →Confidence-Aware Tool Orchestration for Robust Video Understanding▲ 11 on HF · Jun 25, 2026PolicyGuard: A Dialogue-Grounded Sub-Agent Verifier for Policy Adherence in LLM Agents▲ 8 on HF · Jun 28, 2026OmniOPD: Logit-Free On-Policy Distillation via Speculative Verification▲ 8 on HF · May 31, 2026Towards One-to-Many Temporal Grounding▲ 7 on HF · Jun 4, 2026SDR: Set-Distance Rewards for Radiology Report Generation▲ 3 on HF · May 30, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed140 tok/s
Latency (TTFT)18s
Price in / out$1.25 / $10 per 1M
Blended price$3.44 / 1M
Coding index33.3
Math index87.7
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$10/M
Google1M ctx · $10/M