Mercury API
Diffusion models for reasoning, coding, agents and RAG.
Pay as you go; not a monthly subscription.
Official product Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Verified 5 days ago · 2026-10-15
Inception Labs develops diffusion-based language models designed for fast inference, reasoning, coding, and real-time AI applications. Its Mercury model family emphasizes low latency and high throughput.
All plans · 1
API products and access conditions
Usage-based token API. Direct API rates do not establish OpenRouter or other channel prices.
One-time signup credit, not monthly. Official sources disagree on the amount, so no confirmed quantity is published. Docs say payment details are not required. Expiry, regions, enterprise-model eligibility and actual account credit remain unverified.
Publicly documented Inception direct API; account and API key required. Purchase and authenticated inference were not tested. Region eligibility requires confirmation.
Mercury Voice enterprise access
Contact sales@inceptionlabs.ai. LLM text-token fees exclude speech recognition and synthesis.
Official product Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Mercury Router
Model routing, routing analytics and model analytics. Custom pricing; contact sales. Public API access and exact service coverage are unconfirmed.
Custom pricing / Contact Sales. No verified public API identifier or token rate; excluded from model cost rankings.
Contact Sales · Custom Pricing
Official product · Last verified: 2026-10-15T10:02:53.6464958Z
Mercury Decide decisions API
API endpoint: https://api.inceptionlabs.ai/v1/decisions
Input-token billing for classification and scoring; no chat, streaming or tool calls. See the separate model rate record.
Official product Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Legacy Mercury models
Mercury 1 remains supported for existing customers. Current Mercury Coder identity, access and prices remain unverified; no purchasable product is invented.
Price unverified; excluded from lowest-cost rankings.
Official product · Last verified: 2026-10-15T10:02:53.6464958Z
Official billing · Last verified: 2026-10-15T10:02:53.6464958Z
API 产品与使用条件
按量 Token API；官方直连报价不代表 OpenRouter 或其他渠道。
一次性注册赠送，非每月赠送。官网与文档免费 Token 数量有冲突，数量暂不发布；文档说明无需付款信息。到期、地区、企业型号资格及实际到账尚未确认。
官方直连 API 文档公开，需注册账户和 API Key；未执行付费或鉴权推理测试，地区资格待确认。
Mercury API
Diffusion models for reasoning, coding, agents and RAG.
按量使用，非月度订阅。
Mercury Voice 企业服务
联系 sales@inceptionlabs.ai 开通。仅LLM文本Token费用，不含语音识别或合成。
Mercury Router
Model routing, routing analytics and model analytics. Custom pricing; contact sales. Public API access and exact service coverage are unconfirmed.
定制报价，联系销售；公开API及具体服务范围待确认，不填写Token单价。
联系销售 · 定制报价
官方入口 · 最近核验: 2026-10-15T10:02:53.6464958Z
Mercury Decide 决策API
API endpoint: https://api.inceptionlabs.ai/v1/decisions
Input-token billing for classification and scoring; no chat, streaming or tool calls. See the separate model rate record.
历史 Mercury 型号
Mercury 1 仍支持老客户；Mercury Coder 的当前精确型号、访问及价格待确认，未创建可购买商品。
价格待核验，不参与最低成本排序。
官方入口 · 最近核验: 2026-10-15T10:02:53.6464958Z
官方计费 · 最近核验: 2026-10-15T10:02:53.6464958Z
API models and prices · 5
Usage-based billing: text rates are usually per million tokens; other units and tiers are stated per entry. Compare only the same currency, unit and billing conditions. Unverified cache pricing does not mean free.
Mercury 2.5
Current promotional pricing · Rate limits and quotas apply
Type / status: llm / public_api_documented
Use cases: AI Coding, AI Agents, RAG, Low-latency reasoning
API endpoint: https://api.inceptionlabs.ai/v1/chat/completions
Standard and 80%-off promotional rates are separate. Cached input has its own rate; reasoning tokens count as output.
One-time signup credit, not monthly. Official sources disagree on the amount, so no confirmed quantity is published. Docs say payment details are not required. Expiry, regions, enterprise-model eligibility and actual account credit remain unverified.
Publicly documented Inception direct API; account and API key required. Purchase and authenticated inference were not tested. Region eligibility requires confirmation.
Official model ID: mercury-2.5
Input / output: text → text
Model developer: Inception Labs · Service provider: Inception Labs
Model family: — · Currency: USD
| Price dimension | Standard | Current limited-time rate |
|---|---|---|
| Input | USD 0.2 | USD 0.04 |
| Output | USD 0.75 | USD 0.15 |
| Cached read | USD 0.02 | USD 0.004 |
Limited-time offer; the end follows the official announcement. An unknown end date does not mean permanently free.
Estimate text API cost
Inception direct API only; the promotion end date has not been announced.
Documented API context: 260,000 tokens
Maximum output: 65,536 tokens
Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Price verification valid until: 2026-10-22T10:02:53.646Z · This is a verification deadline, not the campaign end date
Mercury Voice
Current promotional pricing · Rate limits and quotas apply
Type / status: llm / enterprise_contact_sales
Use cases: Real-time voice agents
API endpoint: https://api.inceptionlabs.ai/v1/chat/completions
Excluded from default production recommendations; check model availability, pricing and usage conditions.
Text-token LLM inference only. Speech recognition and speech synthesis are separate; these are not end-to-end per-minute voice prices.
One-time signup credit, not monthly. Official sources disagree on the amount, so no confirmed quantity is published. Docs say payment details are not required. Expiry, regions, enterprise-model eligibility and actual account credit remain unverified.
Enterprise customers must contact sales@inceptionlabs.ai for access. The actual API model ID is unverified; access is not assumed for all accounts.
Input / output: text → text
Model developer: Inception Labs · Service provider: Inception Labs
Model family: — · Currency: USD
| Price dimension | Standard | Current limited-time rate |
|---|---|---|
| Input | USD 0.4 | USD 0.2 |
| Output | USD 1.5 | USD 0.75 |
| Cached read | USD 0.04 | USD 0.02 |
Limited-time offer; the end follows the official announcement. An unknown end date does not mean permanently free.
Inception direct API only; the promotion end date has not been announced.
Documented API context: 128,000 tokens
Maximum output: 50,000 tokens
Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Price verification valid until: 2026-10-22T10:02:53.646Z · This is a verification deadline, not the campaign end date
Mercury Decide
Current promotional pricing · Rate limits and quotas apply
Type / status: decision / public_api_documented
Use cases: Classification, Decision scoring, Routing
API endpoint: https://api.inceptionlabs.ai/v1/decisions
Excluded from default production recommendations; check model availability, pricing and usage conditions.
Only input tokens are charged. Output usage is reported but uncharged; this is not a free generative language-model offer.
One-time signup credit, not monthly. Official sources disagree on the amount, so no confirmed quantity is published. Docs say payment details are not required. Expiry, regions, enterprise-model eligibility and actual account credit remain unverified.
Publicly documented decisions API with an account API key; authenticated calls and regional eligibility were not verified.
Official model ID: mercury-decide
Input / output: text, json, image → decision
Model developer: Inception Labs · Service provider: Inception Labs
Model family: — · Currency: USD
| Price dimension | Standard | Current limited-time rate |
|---|---|---|
| Input | USD 0.04 | USD 0.02 |
| Cached read | Unverified | Unverified |
Limited-time offer; the end follows the official announcement. An unknown end date does not mean permanently free.
Inception direct API only; the promotion end date has not been announced.
Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Price verification valid until: 2026-10-22T10:02:53.646Z · This is a verification deadline, not the campaign end date
Mercury 2 · legacy · Historical record
Mercury 2
Paid production API
Type / status: llm / existing_customers_only
Use cases: Legacy customer workloads
API endpoint: https://api.inceptionlabs.ai/v1/chat/completions
Excluded from default production recommendations; check model availability, pricing and usage conditions.
Official direct-channel legacy-model rates observed at the verification time.
One-time signup credit, not monthly. Official sources disagree on the amount, so no confirmed quantity is published. Docs say payment details are not required. Expiry, regions, enterprise-model eligibility and actual account credit remain unverified.
Existing customers only; contact Inception to confirm any new-account access.
Official model ID: mercury-2
Input / output: text → text
Model developer: Inception Labs · Service provider: Inception Labs
Model family: — · Currency: USD
| Price dimension | Current rate |
|---|---|
| Input | USD 0.25 |
| Output | USD 0.75 |
| Cached read | USD 0.025 |
Inception direct rates at verification time, for existing customers only.
Documented API context: 128,000 tokens
Maximum output: 50,000 tokens
Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Mercury Edit 2 · legacy · Historical record
Mercury Edit 2
Paid production API
Type / status: code_edit / existing_customers_only
Use cases: Legacy code editing, Autocomplete
API endpoint: https://api.inceptionlabs.ai/v1/edit/completions
Excluded from default production recommendations; check model availability, pricing and usage conditions.
Text-token billing for FIM/edit endpoints. Tool calling is not assumed.
One-time signup credit, not monthly. Official sources disagree on the amount, so no confirmed quantity is published. Docs say payment details are not required. Expiry, regions, enterprise-model eligibility and actual account credit remain unverified.
Existing customers only; contact sales about continued access or migration.
Official model ID: mercury-edit-2
Input / output: text → text
Model developer: Inception Labs · Service provider: Inception Labs
Model family: — · Currency: USD
| Price dimension | Current rate |
|---|---|
| Input | USD 0.25 |
| Output | USD 0.75 |
| Cached read | USD 0.025 |
Inception direct rates at verification time, for existing customers only.
Documented API context: 32,000 tokens
Maximum output: 8,192 tokens
Official source · Last verified: 2026-10-15T10:02:53.6464958Z
Need a recommendation?
The AI advisor considers your budget, tools, usage and purchase conditions to suggest primary and backup options.
Open AI advisor →