CLOSED TO NEW USERSDIFFUSION LLM260K CONTEXT1,100+ TOKENS/SECACCESS ENDS NOV 15
Mercury 2.5 API Pricing
Mercury 2.5 was Inception's flagship diffusion LLM from September 8, 2026 until Mercury 3 replaced it on October 19. Since then it is closed to new users: accounts that used it before that date keep access until November 15, 2026, and enterprise customers stay on their agreements. The developer table still prices it at $0.20/M input, $0.02/M cached input and $0.75/M output, and the 80% launch discount it shipped with is gone. The context window is 260K.
Input - per 1M tokens
$0.20/M
Launch discount over
Output - per 1M tokens
$0.75/M
Launch discount over
Cached input - per 1M tokens
$0.02/M
90% off fresh input -90%
Effective - agentic blend
$0.108/M
92/8 split - 82% cache
§ 01 / TERMINAL
Mercury 2.5 API cost calculator.
Calculator pre-loaded with Mercury 2.5's rates, which are now what an existing account pays: the 80% launch discount is over. New accounts cannot call this model; run the same workload on Mercury 3's page for the replacement.
$ /mo
Workload split
Prompt cache hit rate
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
§ 02 / SCENARIOS
Mercury 2.5 cost on real workloads.
CODING SUBAGENT
Context compaction
$0.014/task
SEARCH AGENT
Query rewrite and rerank
$0.00382/query
VOICE AGENT
Live call turn
$0.00054/turn
SUPPORT
Ticket reply
$0.00117/ticket
Estimate · inception-tokenizer-estimate · ≈3.85 chars/token Auto-counts as you type
This is a chars-per-token approximation, not a real tokenizer. Actual tokens vary by language, code density, and tool-call overhead — counts are typically ±10–20% off for English prose, more for code or non-Latin scripts. For exact billing, use the vendor's official tokenizer.
Characters 475
Words 80
Tokens (estimated) 123 tokens
Cost as input · uncached $0.00002 USD
Cost as output · uncached $0.00009 USD
Cost as cached input $0.000002 USD
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| Mercury 2.5 Current | $0.20 cache $0.02 | $0.75 | $0.108 agentic 92/8 | 260K | Existing deployments, until November 15, 2026 |
| Mercury 3 | $0.20 cache $0.02 | $0.70 | $0.104 −4% · promo | 1M | The replacement: 1M window, lower output list rate |
| Mercury 2 | $0.25 cache $0.025 | $0.75 | $0.12 deprecated · +11% | 128K | The previous generation, for existing deployments |
| Mercury Edit 2 | $0.25 cache $0.025 | $0.75 | $0.12 +11% | 32K | Autocomplete and next-edit |
| Mercury 2.5 Preview | $0.20 cache $0.02 | $0.75 | $0.108 withdrawn | 260K | The preview this release replaced, for the record |
| GPT-6 Luna | $0.10 cache $0.01 | $0.50 | $0.0682 −37% | 1.05M | High-volume work on a closed frontier stack |
| Claude Haiku 5.5 | $0.10 cache $0.01 | $0.50 | $0.0682 −37% | 1M | Anthropic's fast, high-volume tier |
| Gemini 3.8 Flash | $0.75 cache $0.075 | $3.75 | $0.481 +4.4× · promo | 1M | Google's fast multimodal tier |
| DeepSeek V4.1 Flash | $0.30 cache $0.006 | $1.20 | $0.15 +39% | 1M | Cheap 1M-context agent loops with images |
Mercury 2.5 API pricing FAQ.
What Mercury 2.5 costs now that its launch discount is over, what closing to new users means, and how it compares with Mercury 3 and Mercury 2.
Q · 01 How much does Mercury 2.5 cost? +
$0.20/M input, $0.02/M cached input and $0.75/M output, $0.11/M on our 92/8 agentic blend with 82% cache hits. That is now what an existing account is charged: the 80% launch discount (
$0.04 / $0.004 / $0.15) was still on the developer table on October 6, 2026 and is gone by October 21. Inception never published an end date for it and does not say when it stopped.Q · 02 Can I still use Mercury 2.5? +
Only if your account used it before October 19, 2026. Inception's models page says Mercury 2 and Mercury 2.5 "are no longer available to new users", that existing customers keep access until November 15, 2026, and that enterprise customers remain supported under their agreements. Inception does not say what happens to a call after November 15; plan the move to Mercury 3 before then.
Q · 03 What does moving to Mercury 3 cost? +
Slightly less at list. Mercury 3 lists at the same
$0.20 input and $0.02 cached input and at $0.70 output against this model's $0.75, with a 1M window against 260K and up to 131,072 output tokens. During Mercury 3's own launch discount it costs half of that. The id changes from mercury-2.5 to mercury-3 on the same v1/chat/completions endpoint.Q · 04 What changed from Mercury 2.5 Preview? +
Where you called it, not what it cost. The preview was listed at the same
$0.20 / $0.02 / $0.75 but served only through OpenRouter. Mercury 2.5 moved to Inception's own API under the id mercury-2.5 on September 8, 2026, with an 80% launch discount that has since ended.Q · 05 How does it compare with Mercury 2? +
It is cheaper on input and cached input (
$0.20 and $0.02 against $0.25 and $0.025), identical on output, with twice the context (260K against 128K) and a higher output ceiling (65,536 against 50,000 tokens). Inception claims a 40% gain in intelligence over Mercury 2. Both closed to new users on the same day and share the November 15, 2026 cut-off.Q · 06 How fast is it, and does speed change the bill? +
Inception quotes 1,107 tokens per second on widely available NVIDIA GPUs; that is the vendor's figure, not our measurement. Billing is per token as usual, so speed does not lower the token bill. It changes what latency-bound work costs in time: Inception's own examples are search pipelines that make dozens of model calls per user request, voice agents answering in under 200 milliseconds and context compaction inside coding agents.
Q · 07 Where can I use it? +
On Inception's own OpenAI-compatible API for accounts that already used it, and through Baseten and OpenRouter, where a reseller sets its own margin and its own schedule. The rates on this page are Inception's own. The model takes text in and gives text out, with tunable reasoning, parallel tool calls and schema-aligned JSON output.