GPT-5.6 (Sol / Terra / Luna) is now evaluated on TrustVector โ€” with day-1 independent verification, incl. METR's benchmark-cheating findings.

Read the evaluation
Evaluation record ยท gemini-2-5-pro

Gemini 2.5 Pro

vgemini-2.5-pro-002

Google

Modeldeprecatedlong-context1m-tokensdeep-think
89
Strong
About This Model

DEPRECATED: Google has scheduled Gemini 2.5 Pro for shutdown on 2026-10-16; the designated replacement is Gemini 3.1 Pro. Formerly Google's flagship (superseded by the Gemini 3.x line since late 2025), with 1M token context window (2M on select Vertex AI enterprise tiers), Deep Think mode for complex reasoning, and native multimodal capabilities. Do not start new projects on this model.

Last Evaluated: July 9, 2026
Official Website

Trust Vector Analysis

Dimension Breakdown

๐Ÿš€Performance & Reliability
+

Strong performance for its generation with 1M context window and Deep Think mode, but superseded by the Gemini 3.x line. DEPRECATED: shutdown scheduled 2026-10-16.

task accuracy code

Standard coding benchmarks

Evidence
HumanEval โ€” 88.4% on HumanEval
highVerified: 2026-07-09
task accuracy reasoning

PhD-level reasoning benchmarks

Evidence
GPQA Diamond โ€” 69.8% with Deep Think mode
highVerified: 2026-07-09
task accuracy general

Comprehensive knowledge testing

Evidence
MMLU-Pro โ€” 79.1% on graduate knowledge
highVerified: 2026-07-09
output consistency

Internal consistency testing

Evidence
Google AI Documentation โ€” Consistent outputs across requests
mediumVerified: 2026-07-09
latency p50

API latency measurements

Evidence
Community benchmarking โ€” Median latency ~1.5s
mediumVerified: 2026-07-09
latency p95

95th percentile measurements

Evidence
Community benchmarking โ€” p95 latency ~3.5s
mediumVerified: 2026-07-09
context window

Official specification

Evidence
Google AI Documentation โ€” 1M token context window on the standard Gemini API; 2M extension limited to select Vertex AI enterprise tiers and never rolled out to the standard API
highVerified: 2026-07-09
uptime

Historical uptime data

Evidence
Google Cloud Status โ€” 99.9% uptime (last 90 days)
highVerified: 2026-07-09
๐Ÿ›ก๏ธSecurity
+

Strong security leveraging Google Cloud infrastructure. Configurable safety filters provide flexibility.

prompt injection resistance

OWASP LLM security testing

Evidence
Google AI Safety โ€” Strong prompt injection defenses
mediumVerified: 2026-07-09
jailbreak resistance

Adversarial prompt testing

Evidence
Community testing โ€” Good resistance to jailbreak attempts
mediumVerified: 2026-07-09
data leakage prevention

Privacy policy review

Evidence
Google Privacy Policy โ€” Data handling policies for Gemini API
mediumVerified: 2026-07-09
output safety

Safety testing

Evidence
Google Safety Filters โ€” Configurable safety filters across categories
highVerified: 2026-07-09
api security

API security review

Evidence
Google Cloud Security โ€” Google Cloud security standards
highVerified: 2026-07-09
๐Ÿ”’Privacy & Compliance
+

Good privacy with Google Cloud infrastructure. Enterprise options provide enhanced controls. HIPAA compliance available through Google Cloud.

data residency

Cloud infrastructure review

Evidence
Google Cloud Regions โ€” Multiple region options via Google Cloud
highVerified: 2026-07-09
training data optout

Terms review

Evidence
Gemini API Terms โ€” API data not used for training
highVerified: 2026-07-09
data retention

Data retention policy review

Evidence
Google Cloud Data Handling โ€” Retention policies vary by service tier
mediumVerified: 2026-07-09
pii handling

Data protection review

Evidence
Google AI Safety โ€” Customer responsible for PII handling
mediumVerified: 2026-07-09
compliance certifications

Certification verification

Evidence
Google Cloud Compliance โ€” SOC 2, ISO 27001, GDPR compliant
highVerified: 2026-07-09
zero data retention

Enterprise feature review

Evidence
Enterprise Options โ€” Available for enterprise customers
mediumVerified: 2026-07-09
๐Ÿ‘๏ธTrust & Transparency
+

Strong transparency with Deep Think mode and comprehensive documentation. Configurable guardrails provide flexibility.

explainability

Reasoning transparency evaluation

Evidence
Deep Think Feature โ€” Deep Think mode exposes reasoning process
highVerified: 2026-07-09
hallucination rate

Factual QA testing

Evidence
Google AI Testing โ€” Improved factual accuracy over Gemini 1.5
mediumVerified: 2026-07-09
bias fairness

Bias benchmark evaluation

Evidence
Google AI Principles โ€” Regular bias testing and mitigation
mediumVerified: 2026-07-09
uncertainty quantification

Qualitative assessment

Evidence
Model behavior โ€” Expresses uncertainty appropriately
mediumVerified: 2026-07-09
model card quality

Documentation review

Evidence
Gemini Model Card โ€” Comprehensive model documentation
highVerified: 2026-07-09
training data transparency

Public disclosure review

Evidence
Google AI Blog โ€” General training data description
mediumVerified: 2026-07-09
guardrails

Safety mechanism review

Evidence
Safety Settings โ€” Configurable multi-category safety filters
highVerified: 2026-07-09
โš™๏ธOperational Excellence
+

Excellent operational maturity backed by Google Cloud infrastructure, but the model is deprecated with a 2026-10-16 shutdown date. Versioning score reduced to reflect the pending retirement; migrate to Gemini 3.1 Pro.

api design quality

API design review

Evidence
Gemini API โ€” RESTful API with streaming, function calling, native multimodal
highVerified: 2026-07-09
sdk quality

SDK quality assessment

Evidence
Google AI SDKs โ€” SDKs for Python, Node.js, Go, Swift, Kotlin
highVerified: 2026-07-09
versioning policy

Versioning policy review

Evidence
Google Cloud Versioning โ€” Clear versioning with migration guides
Gemini API Deprecations โ€” gemini-2.5-pro scheduled for shutdown 2026-10-16 (extended from an original June 2026 date); designated replacement is gemini-3.1-pro
highVerified: 2026-07-09
monitoring observability

Observability tools review

Evidence
Google Cloud Console โ€” Comprehensive Cloud Console monitoring
highVerified: 2026-07-09
support quality

Support assessment

Evidence
Google Cloud Support โ€” Enterprise support with SLAs
highVerified: 2026-07-09
ecosystem maturity

Ecosystem analysis

Evidence
Google AI Ecosystem โ€” Growing ecosystem with Google Cloud integration
highVerified: 2026-07-09
license terms

License review

Evidence
Google Cloud Terms โ€” Standard commercial terms
highVerified: 2026-07-09
Strengths
  • +1M token context window (2M on select Vertex AI enterprise tiers)
  • +Deep Think mode for enhanced reasoning on complex problems
  • +Native multimodal capabilities (text, image, video, audio)
  • +Google Cloud infrastructure with enterprise-grade reliability
  • +Excellent for massive document analysis and research
  • +Competitive pricing with strong performance
Limitations
  • !Slightly behind Claude/GPT on specialized benchmarks
  • !Deep Think mode increases latency significantly
  • !Data retention policies less transparent than Anthropic
  • !Smaller ecosystem than OpenAI
  • !Superseded by the Gemini 3.x line (Gemini 3.1 Pro is the current Pro tier)
  • !DEPRECATED: shutdown scheduled 2026-10-16; migrate to Gemini 3.1 Pro
Metadata
pricing
input: $1.25 per 1M tokens (<200k context), $2.50 per 1M tokens (>200k context)
output: $10.00 per 1M tokens (<200k context), $15.00 per 1M tokens (>200k context)
notes: Tiered pricing based on context length. Confirmed unchanged on the official pricing page as of 2026-07-09; billing ends when the model shuts down 2026-10-16.
last verified: 2026-07-09
context window: 1000000
languages
0: English
1: 100+ languages
modalities
0: text
1: vision
2: audio
3: video
api endpoint: https://generativelanguage.googleapis.com/v1beta/models
open source: false
architecture: Multimodal transformer with Deep Think reasoning
parameters: Not disclosed

Use Case Ratings

code generation

Strong coding capabilities. Excellent for code explanation and documentation with long context.

customer support

Good for customer support with multimodal capabilities. Can process images and documents natively.

content creation

Excellent for content creation with good creativity and natural writing.

data analysis

Outstanding for data analysis with 1M context enabling analysis of massive datasets.

research assistant

Exceptional for research with 1M context. Can process entire books, papers, and repositories.

legal compliance

Good for legal work with massive context enabling full contract analysis.

healthcare

Good capabilities with HIPAA compliance available via Google Cloud. Large context useful for medical records.

financial analysis

Strong for financial analysis with ability to process large financial documents.

education

Excellent for education with multimodal capabilities and patient explanations.

creative writing

Good for creative writing with strong narrative capabilities.