Mercury 2.5 API

Inception's diffusion LLM: the fastest reasoning model in production at 1,107 tokens/sec.

Modalities: Image, TextText
In / Out price: $0.2116/$0.7935per 1M
Context window: 260K tokens
Released: Sep 8, 2026

Mercury 2.5 overview

Mercury 2.5 by Inception

Mercury 2.5 is Inception's diffusion language model, generating and refining many tokens in parallel rather than one after another.

Mercury 2.5 specs

Model ID: inception/mercury-2.5
Developer: Inception
Context window: 260,000 tokens
Max output: 65,536 tokens
Input: Image, Text
Output: Text
API: OpenAI-compatible, base URL https://api.aimlapi.com/v1

What Is Mercury 2.5?

Mercury 2.5 is the largest diffusion language model trained to date and, per Inception, the fastest reasoning LLM in production. Instead of emitting tokens one at a time, it produces and refines many tokens in parallel, the core idea behind diffusion-based language models.

Inception reports a 40% increase in intelligence over Mercury 2 and 1,107 tokens per second on widely available NVIDIA GPUs, placing it alongside cost-optimised frontier models on quality while running considerably faster.

Technical Specifications

Context window: 260,000 tokens
File input, vision, web search, parallel tool calls and structured output

Where Mercury 2.5 Fits

Latency-sensitive agents
Throughput above 1,000 tokens/sec makes multi-step agent loops feel interactive rather than batched.

High-volume generation
Parallel decoding keeps cost per token low at scale.

Released by Inception on August 31, 2026 and available on AI/ML API since September 8, 2026.

Use cases

Mercury 2.5 is documented for fast chat, agents, and structured workflows. It is also suited to search and RAG pipelines that make many calls per request.

It is documented for voice agents and voice pipelines. Inception also lists Mercury Voice use cases including customer support, patient care, education, and gaming, though those uses are shown for the voice product line rather than Mercury 2.5 specifically.

It is documented for coding-assistant tasks, including context compaction, routing, autocomplete, and next-edit suggestions. Inception also says it serves latency-sensitive coding products and coding subagents.

Capabilities

File input
Reasoning
Streaming
Structured outputs
Function calling
Vision
Web search

Mercury 2.5 benchmarks by Artificial Analysis

See how Mercury 2.5 scores on independent evaluations. Data retrieved Nov 3, 2026.

Humanity's Last Exam#193 / 45811.8%Score Terminal-Bench 4.0#88 / 1540%Score AA-LCR#128 / 39171.7%Score SciCode#115 / 15938.5%Score

Benchmark results

Compare 12 models available on AI/ML API across 4 independent benchmarks.

Humanity's Last Exam

Expert-written questions across academic subjects.

Mercury 2.5 API pricing

Mercury 2.5 pricing in US dollars
Usage	Price	Unit
Input	$0.2116	1M tokens
Output	$0.7935	1M tokens
Cached input	$0.0212	1M tokens

Prices from the AI/ML API price list, checked on Nov 4, 2026.

Mercury 2.5 vs other models: token prices

AI/ML API input and output rates for 5 models, per million tokens.

AI/ML API model prices in USD per million tokens
Model	Input	Output
Mercury 2.5	$0.2116	$0.7935
Mercury 3	$0.2116	$0.7406
Mercury 2	$0.2645	$0.7935
Small 4	$0.1984	$0.7935
ReMM-SLERP-L2-13B	$0.529	$0.7935

How to use the Mercury 2.5 API

View docs

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/chat/completions' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "inception/mercury-2.5",
  "messages": [
    {
      "role": "user",
      "content": "Write a haiku about the ocean."
    }
  ]
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "inception/mercury-2.5",
    "messages": [
        {
            "role": "user",
            "content": "Write a haiku about the ocean."
        }
    ]
}

response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "inception/mercury-2.5",
    "messages": [
      {
        "role": "user",
        "content": "Write a haiku about the ocean."
      }
    ]
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

All posts

Mercury 2.5 FAQ

What is Mercury 2.5?

Mercury 2.5 is a diffusion language model from Inception. Unlike models that emit one token after another, it generates and refines many tokens in parallel.

How does a diffusion language model differ from a standard LLM?

A standard LLM predicts tokens one at a time, each conditioned on the previous one. A diffusion language model starts from a rough draft of the whole output and refines it over several passes, so many tokens improve at once.

How fast is Mercury 2.5?

Inception reports throughput of 1,107 tokens per second on widely available NVIDIA GPUs, and describes it as the fastest reasoning model in production.

Does Mercury 2.5 support tool calling?

Yes. It supports function calling and can issue parallel tool calls within a single turn.

Can Mercury 2.5 return structured JSON?

Yes. It supports schema-aligned JSON output, so responses can be constrained to the shape your code expects.

Can you control how much Mercury 2.5 reasons?

Yes. Reasoning depth is tunable, so you can trade latency against deliberation on a per-request basis.

What inputs and outputs does Mercury 2.5 support?

Mercury 2.5 takes text and image input and returns text. It also supports file input and web search.

How do I call Mercury 2.5 on AI/ML API?

Send a request to the chat completions endpoint with the model id inception/mercury-2.5. The request shape is the same as for any other chat model on AI/ML API.

How does Mercury 2.5 compare with Mercury 2?

Inception reports roughly a 40% increase in intelligence over Mercury 2, while keeping the parallel decoding that makes the family fast.

What is Mercury 2.5 best suited for?

Latency-sensitive work such as agent loops that make many sequential calls, and high-volume generation where throughput is the constraint.

Start building with Mercury 2.5

Get API key

Mercury 2.5 API

inception/mercury-2.5

Build with your agent

Give your agent the model details, code and documentation.

1Get your API key

Create API key

export AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"

2Make your first request

curl 'https://api.aimlapi.com/v1/chat/completions' \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H 'Content-Type: application/json' \
  -d '{
  "model": "inception/mercury-2.5",
  "messages": [
    {
      "role": "user",
      "content": "Write a haiku about the ocean."
    }
  ]
}'

import os
import requests

headers = {
    "Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
    "Content-Type": "application/json",
}
payload = {
    "model": "inception/mercury-2.5",
    "messages": [
        {
            "role": "user",
            "content": "Write a haiku about the ocean."
        }
    ]
}

response = requests.post("https://api.aimlapi.com/v1/chat/completions", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())

const response = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "inception/mercury-2.5",
    "messages": [
      {
        "role": "user",
        "content": "Write a haiku about the ocean."
      }
    ]
  }),
});

if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());

Docs for AI agents Plain-text index · llms.txt API documentation API reference and examples

Talk to our sales team

Thanks for reaching out

Mercury 2.5 API

Mercury 2.5 overview

Mercury 2.5 by Inception

Mercury 2.5 specs

What Is Mercury 2.5?

Technical Specifications

Where Mercury 2.5 Fits

Use cases

Capabilities

Mercury 2.5 benchmarks by Artificial Analysis

Benchmark results

Humanity's Last Exam

Terminal-Bench 4.0

AA-LCR

SciCode

Mercury 2.5 API pricing

Mercury 2.5 vs other models: token prices

How to use the Mercury 2.5 API

Build with your agent

1Get your API key

2Make your first request

Related articles

Mercury 2.5 FAQ

Start building with Mercury 2.5

Mercury 2.5 API

Build with your agent

1Get your API key

2Make your first request

Mercury 2.5 API

Mercury 2.5 overview

Mercury 2.5 by Inception

Mercury 2.5 specs

What Is Mercury 2.5?

Technical Specifications

Where Mercury 2.5 Fits

Use cases

Capabilities

Benchmark results

Humanity's Last Exam

Terminal-Bench 4.0

AA-LCR

SciCode

Mercury 2.5 API pricing

Mercury 2.5 vs other models: token prices

How to use the Mercury 2.5 API

Build with your agent

1Get your API key

2Make your first request

Models similar to Mercury 2.5

Mercury 3

Mercury 2

Hy3

Nemotron 3.5 Content Safety

GPT-5.4 Nano

GPT-5.6 Luna Pro

Related articles

Best AI Models for Agentic Workflows and Tool Use in 2026

Best LLMs for Long-Context & Multimodal Tasks in 2026

Mercury 2.5 FAQ

Start building with Mercury 2.5