Inception · Balanced
Mercury 2
Inception Labs' fastest reasoning model with tool calling and structured outputs support.
- Reasoning
- Tools
- Structured output
- Input
- Text
- Output
- Text
- Discovery tier
- Balanced
Pricing
USD per 1M tokens- Input
- $0.25
- Output
- $0.75
- Cache read
- $0.025
Reference prices as published by the model provider. API example
Specifications
- Context window
- 128K
- Max output
- 50K
Side by side
Mercury 2 compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Mercury 2InceptionThis model | Mercury 3InceptionRemove | |
|---|---|---|
| Provider | Inception | Inception |
| Discovery tier | Balanced | Efficient |
| Added | Jan 1, 2024 | Oct 19, 2026 |
| Limits | ||
| Context window | 128K | 1MBest |
| Max output | 50K | 131.07KBest |
| Average output | — | 1.9K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.25 | $0.10Best |
| Output | $0.75 | $0.35Best |
| Cache read | $0.025 | $0.01Best |
| Performance | ||
| Throughput | — | 913.7 tok/s |
| Time to first token | — | 1.9 s |
| Request success (24h) | — | 100.0% |
| Benchmark indexes | ||
| Intelligence Index | 10.8 | — |
| Coding Index | 31.1 | — |
| Capabilities | ||
| Reasoning | Yes | Yes |
| Tools | Yes | Yes |
| Structured output | Yes | Yes |
| Parallel tools | No | Yes |
| Reasoning effort | — | none, low, medium, high |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Mercury 2
- Mercury 3
Intelligence Index
- Mercury 210.8
- Mercury 3—
Mercury 2: Better than 23% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Mercury 2: 31.1
- Mercury 3: —
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Mercury 2: 1.7%
- Mercury 3: —
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Mercury 2: 0.3%
- Mercury 3: —
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Mercury 2: 265 Elo
- Mercury 3: —
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Mercury 2: 494 Elo
- Mercury 3: —
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Mercury 2: 3.8%
- Mercury 3: —
AA-LCRv1.1
Reasoning across long documents
- Mercury 2: 43.7%
- Mercury 3: —
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Mercury 2: 17.1%
- Mercury 3: —
GPQA Diamond
Graduate-level science questions
- Mercury 2: 77.0%
- Mercury 3: —
IFBench
Following precise output instructions
- Mercury 2: 69.8%
- Mercury 3: —
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Mercury 2: 0.0%
- Mercury 3: —
SciCode
Scientific computing in Python
- Mercury 2: 37.7%
- Mercury 3: —
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- Mercury 2: 26.5%
- Mercury 3: —
τ²-Bench Telecom
Tool use in a customer-support simulation
- Mercury 2: 70.8%
- Mercury 3: —
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Mercury 2: 43.7%
- Mercury 3: —
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
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": "mercury-2",
"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="mercury-2",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Inception
Related models
Other Inception models, starting with the same family.