impulse

Mercury 2.5

API id mercury-2.5
Input / 1M
$0.20
at Inception
Output / 1M
$0.75
at Inception
Context window
260K
tokens
Max output
66K
tokens per reply
Hosts
1
1 listings
Released
-
date not published

Pricing

List prices at Inception, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.20$0.75
Cached input
90% off input
$0.02

What it costs a month

At Inception list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$100
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$79.80
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$134
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$660
Use your own numbers

Where to run it

1 host serve Mercury 2.5. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
mercury-2.5
$0.20 / $0.75

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Features
Prompt caching

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than Inception.

Smaller Independent models

No cheaper Independent model keeps its key capabilities.

Other frontier labs

  • Claude Haiku 5.5
    Anthropic · on Anthropic
    $0.10 / $0.50
    Has every key capability
    41% less
  • GPT-6 Luna
    OpenAI · on OpenAI
    $0.10 / $0.50
    Has every key capability
    41% less
  • GPT-4o Mini
    OpenAI · on OpenAI
    $0.15 / $0.60
    Has every key capability
    22% less

Open-weight picks

  • Deepseekv4.1 Flash
    DeepSeek · on Sarvam
    $0.149 / $0.585
    Has every key capability
    24% less
  • Qwen3.8 27B
    Alibaba Qwen · on EmpirioLabs AI
    $0.17 / $0.50
    Has every key capability
    25% less
  • Nemotron 3.5 Super VL
    NVIDIA · on Kyma API
    $0.081 / $0.472
    Has every key capability
    47% less
  • MiMo V2.6 Flash
    Xiaomi · on GMI Cloud
    $0.112 / $0.224
    Has every key capability
    59% less

Model line

Newer and older versions, at each lab's own price.

  1. UndatedMercury 2.5Latest$0.20 / $0.75
  2. UndatedMercury 2$0.25 / $0.75

Call it

Using Inception's API.

Python
from litellm import completion

response = completion(
    model="inception/mercury-2.5",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Mercury 2.5 cost?

On Inception, Mercury 2.5 costs $0.20 per million input tokens and $0.75 per million output tokens. Cached input costs $0.02 per million tokens.

Which APIs offer Mercury 2.5?

Mercury 2.5 runs on 1 API: Inception.

What is Mercury 2.5's context window?

260K tokens, with up to 66K tokens in one reply.

Is Mercury 2.5 being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Mercury 2.5?

No. Mercury 2.5 is the newest model in its line.

What are cheaper alternatives to Mercury 2.5?

Models that cost less and keep most of what it can do: Claude Haiku 5.5 on Anthropic (41% less), GPT-6 Luna on OpenAI (41% less) and Deepseekv4.1 Flash on Sarvam (24% less). Savings are on a mix of 3 input tokens to 1 output token.

What is Mercury 2.5's API model id?

Use mercury-2.5 with Inception.

Sources

Raw catalog entry for mercury-2.5 on Inception
{
  "cache_read_input_token_cost": 2e-8,
  "input_cost_per_token": 2e-7,
  "litellm_provider": "inception",
  "max_input_tokens": 260000,
  "max_output_tokens": 65536,
  "mode": "chat",
  "output_cost_per_token": 7.5e-7,
  "source": "https://docs.inceptionlabs.ai/get-started/models",
  "supports_function_calling": true,
  "supports_prompt_caching": true,
  "supports_response_schema": true,
  "supports_system_messages": true,
  "supports_tool_choice": true
}