Google · Balanced
Gemini 2.5 Pro Preview 0506
Gemini 2.5 Pro Preview 0506. Google's model from May 2025.
- Reasoning
- Vision
- Audio input
- Input
- Text, Images, Audio
- Output
- Text
- Reasoning effort
- low · medium · high
- Discovery tier
- Balanced
Pricing
USD per 1M tokens- Input
- $2.50
- Output
- $10.00
- Cache read
- $0.25
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1M
- Max output
- 65.5K
Side by side
Gemini 2.5 Pro Preview 0506 compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Gemini 2.5 Pro Preview 0506GoogleThis model | Gemini 2.5 ProGoogleRemove | |
|---|---|---|
| Provider | ||
| Discovery tier | Balanced | Frontier |
| Added | May 6, 2025 | Jun 5, 2025 |
| Limits | ||
| Context window | 1.05M | 1.05M |
| Max output | 65.54K | 65.54K |
| Average output | — | 1.1K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $2.50 | $1.25Best |
| Output | $10.00 | $10.00 |
| Cache read | $0.25 | $0.13Best |
| Cache write | — | $0.38 |
| Performance | ||
| Throughput | — | 189.7 tok/s |
| Time to first token | — | 12.3 s |
| Request success (24h) | — | 99.5% |
| Benchmark indexes | ||
| Intelligence Index | 16.1 | 16.1 |
| Coding Index | 33.3 | 33.3 |
| Math Index | 87.7 | 87.7 |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | No | Yes |
| Vision | Yes | Yes |
| Audio input | Yes | Yes |
| Video input | No | Yes |
| Reasoning effort | low, medium, high | low, medium, high |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Gemini 2.5 Pro Preview 0506
- Gemini 2.5 Pro
Intelligence Index
- Gemini 2.5 Pro Preview 050616.1
- Gemini 2.5 Pro16.1
Gemini 2.5 Pro Preview 0506: Better than 41% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Gemini 2.5 Pro Preview 0506: 33.3Top
- Gemini 2.5 Pro: 33.3Top
Math Index
Composite of the math evaluations
- Gemini 2.5 Pro Preview 0506: 87.7Top
- Gemini 2.5 Pro: 87.7Top
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Gemini 2.5 Pro Preview 0506: 2.2%Top
- Gemini 2.5 Pro: 2.2%Top
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Gemini 2.5 Pro Preview 0506: 0.0%Top
- Gemini 2.5 Pro: 0.0%Top
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Gemini 2.5 Pro Preview 0506: 302 EloTop
- Gemini 2.5 Pro: 302 EloTop
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Gemini 2.5 Pro Preview 0506: 455 EloTop
- Gemini 2.5 Pro: 455 EloTop
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Gemini 2.5 Pro Preview 0506: 10.2%Top
- Gemini 2.5 Pro: 10.2%Top
AA-LCRv1.1
Reasoning across long documents
- Gemini 2.5 Pro Preview 0506: 69.0%Top
- Gemini 2.5 Pro: 69.0%Top
MLCR-AA
Long-context reasoning over medical records
- Gemini 2.5 Pro Preview 0506: 12.2%Top
- Gemini 2.5 Pro: 12.2%Top
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Gemini 2.5 Pro Preview 0506: 22.5%Top
- Gemini 2.5 Pro: 22.5%Top
GPQA Diamond
Graduate-level science questions
- Gemini 2.5 Pro Preview 0506: 84.4%Top
- Gemini 2.5 Pro: 84.4%Top
MMLU-Pro
Multi-subject knowledge and reasoning
- Gemini 2.5 Pro Preview 0506: 86.2%Top
- Gemini 2.5 Pro: 86.2%Top
IFBench
Following precise output instructions
- Gemini 2.5 Pro Preview 0506: 48.7%Top
- Gemini 2.5 Pro: 48.7%Top
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Gemini 2.5 Pro Preview 0506: 0.0%Top
- Gemini 2.5 Pro: 0.0%Top
SciCode
Scientific computing in Python
- Gemini 2.5 Pro Preview 0506: 46.3%Top
- Gemini 2.5 Pro: 46.3%Top
LiveCodeBench
Recent competitive programming problems
- Gemini 2.5 Pro Preview 0506: 80.1%Top
- Gemini 2.5 Pro: 80.1%Top
Math
AIME 2025
Competition mathematics, 2025 problems
- Gemini 2.5 Pro Preview 0506: 87.7%Top
- Gemini 2.5 Pro: 87.7%Top
AIME
Competition mathematics
- Gemini 2.5 Pro Preview 0506: 88.7%Top
- Gemini 2.5 Pro: 88.7%Top
MATH-500
Competition math problems
- Gemini 2.5 Pro Preview 0506: 96.7%Top
- Gemini 2.5 Pro: 96.7%Top
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- Gemini 2.5 Pro Preview 0506: 26.5%Top
- Gemini 2.5 Pro: 26.5%Top
τ²-Bench Telecom
Tool use in a customer-support simulation
- Gemini 2.5 Pro Preview 0506: 54.1%Top
- Gemini 2.5 Pro: 54.1%Top
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Gemini 2.5 Pro Preview 0506: 69.0%Top
- Gemini 2.5 Pro: 69.0%Top
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
- Gemini 2.5 Pro Preview 0506
- Gemini 2.5 Pro
| Leaderboard | Gemini 2.5 Pro Preview 0506 | Gemini 2.5 Pro |
|---|---|---|
| Vision Style Control · Captioning | Not in top 20 | #3 · 1250.6 |
| Vision Style Control · Entity Recognition | Not in top 20 | #5 · 1250.8 |
Ranks among Noumen catalog models on LMArena leaderboards. Only top-20 placements are listed.LMArena
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-2.5-pro-preview-05-06",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="gemini-2.5-pro-preview-05-06",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Related models
Other Google models, starting with the same family.
