Monid
Marketplace and payment layer for AI agent tools, enabling discovery, comparison, and pay-per-use API access.
Website: https://monid.ai/
Cover Block
Open sources
| Field | Detail |
|---|---|
| Company | Monid |
| Tagline | Marketplace and payment layer for AI agent tools, enabling discovery, comparison, and pay-per-use API access [fundup.ai] |
| Headquarters | San Francisco, California, United States [Crunchbase] |
| Stage | Seed [Dealroom News, October 2026] |
| Business model | API / Developer Platform |
| Industry | Deeptech |
| Technology | AI / Machine Learning |
| Geography | North America |
| Growth profile | Venture Scale |
| Founding team | Co-Founders (2), Shengkun Ye and Feiyou Guo [chatgate.ai] |
| Funding label | Seed |
| Total disclosed funding | Approximately $9.8 million, based on a reported $2.1 million pre-seed in September 2026 and a reported $7.7 million seed in October 2026 [trysignalbase.com, September 2026] [Dealroom News, October 2026] |
Links
Open sources
- Website: https://monid.ai/docs
- LinkedIn: https://www.linkedin.com/company/monid-ai
What an Investor Needs First
PUBLIC Monid is building a marketplace and payment layer for AI agent tools, and it merits attention now because public reporting suggests the company moved from launch to a catalog of more than 1,700 endpoints within months, alongside fresh seed financing [Dealroom News, October 2026] [Crunchbase]. The public record is still thin, but the core proposition is straightforward: one integration lets an AI agent discover, inspect, run, and pay for third-party APIs on a pay-per-call basis rather than through separate vendor contracts and billing relationships [The Neuron, September 2026] [monid.ai].
The founding story is only partially visible from public sources. Monid is based in San Francisco and appears to have been co-founded by Shengkun Ye and Feiyou Guo, with the product launched in April 2026 at roughly 200 tools before scaling reported coverage to more than 1,700 endpoints by September 2026 [Crunchbase] [chatgate.ai] [Dealroom News, October 2026]. That early expansion is the clearest reason investors are paying attention, even if several usage figures still rest on company-linked or lightly corroborated sources.
Product differentiation, as publicly described, sits less in frontier model development and more in orchestration and commerce. Monid has been described as an "OpenRouter for agent tools," with one shared balance and one access layer across a large catalog of APIs, and reports indicate the platform was open-sourced under an MIT license while offering access to 2,000+ endpoints from 70+ providers [The Neuron, September 2026]. If that positioning holds, the bet is that agent developers value simpler discovery, unified payments, and lower switching costs more than direct point integrations.
The team evidence is mixed but directionally relevant. Public profiles tie Ye to prior product work at TabaPay, and secondary profiles associate him with Founders, Inc., Deserve, and WeChat, while Guo is publicly linked to Founders, Inc. support [Crunchbase] [ZoomInfo] [rocketreach.co] [f.inc]. For this company, that matters because a payment layer for autonomous software agents needs product judgment at the intersection of developer workflow and transaction infrastructure.
On capital formation, the disclosed financing totals about $9.8 million across a reported $2.1 million pre-seed in September 2026 and a $7.7 million seed in October 2026, with investors reported to include 1984 Ventures, Llama Ventures, Untapped Capital, Founders, Inc., Long Journey Ventures, Madrona, Essence VC, and Pioneer Fund [trysignalbase.com] [Dealroom News, October 2026] [Crunchbase]. The business model is also unusually clear for a young infrastructure company: Monid's own documentation says users pay only for what they use, with no subscriptions or contracts, which aligns the company with transaction volume rather than seat expansion [monid.ai].
Over the next 12 to 18 months, the key questions are whether reported transaction volume converts into durable developer adoption, whether the catalog can expand without quality or reliability problems, and whether Monid can become a trusted payment and routing layer rather than a temporary aggregation interface [Shengkun Ye LinkedIn] [en.wowtale.net, September 2026] [The Neuron, September 2026]. Public traction figures include more than 4 million agent transactions, but independent customer validation remains limited, so the near-term watch item is proof that usage depth, provider breadth, and monetization scale together rather than move independently [en.wowtale.net, September 2026] [fundup.ai].
Reasoned from indirect evidence -- This section relies on a mix of named-publisher reporting, Crunchbase, company materials, and secondary profiles; several material operating metrics remain only partially corroborated.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Seed |
| Business Model | API / Developer Platform |
| Industry / Vertical | Deeptech |
| Technology Type | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (2) |
| Funding | Seed, total disclosed about $9.8 million |
Inside the Company
PUBLIC
The public record on Monid is thin, but the company presents clearly enough at the operating level: it is a San Francisco, California startup building a single access layer for AI agents to discover and use third-party tools and APIs [Crunchbase]. On its documentation site, Monid describes a pay-per-use model in which users can access endpoints on demand without subscriptions or contracts, which aligns with Crunchbase's summary of the product as a unified registry and access layer for agent tools [monid.ai] [Crunchbase].
What is dated with more confidence is the sequence of visible milestones rather than the founding date. Dealroom News reported that Monid launched in April 2026 with roughly 200 tools and had expanded to more than 1,700 endpoints by September 2026 [Dealroom News, October 2026]. Crunchbase lists Shengkun Ye and Feiyou Guo in connection with the company, and names Ye as co-founder on his person profile, but the legal entity name and incorporation details were not confirmed in the cited public sources used for this section [Crunchbase].
Partially corroborated -- Core company identity is supported by Crunchbase and the company website, but chronology in this section relies in part on a single named-publisher report.
Under the Hood
MIXED
The core product appears to be a transaction layer for AI agents rather than a single-purpose tool. Public descriptions are fairly consistent on the basic workflow: Monid says developers can give an agent one integration point to discover, inspect, call, and pay for third-party APIs on demand, with usage billed from a shared prepaid balance rather than through separate vendor contracts [monid.ai] [The Neuron, September 2026]. Crunchbase uses similar language, describing Monid as an all-in-one tool registry that connects AI agents to more than 1,700 tools and APIs through a single integration [Crunchbase].
The public record also suggests the company is trying to solve two friction points at once, discovery and checkout. Dealroom reported that Monid launched in April 2026 with roughly 200 tools and expanded to more than 1,700 endpoints by September 2026 [Dealroom News, October 2026]. The Neuron separately described the product as an MIT-licensed "OpenRouter for agent tools" and said it offered 2,000+ endpoints from 70+ providers as of September 2026 [The Neuron, September 2026]. Those figures are directionally aligned on rapid catalog expansion, but they do not fully match, so the safer read is that Monid had reached the high hundreds to low thousands of accessible endpoints by late 2026 across a multi-provider network [Dealroom News, October 2026] [The Neuron, September 2026].
The technical differentiation that can be discussed publicly is therefore more architectural than model-centric. Monid's own documentation emphasizes on-demand access to hundreds of web data endpoints and a pay-per-use model with no subscriptions or contracts [monid.ai]. Press coverage and company materials both frame the product as a single access and payment layer for agentic workflows, which, if durable, could reduce integration overhead for developers building agents that need many external tools rather than one deeply embedded API [The Neuron, September 2026] [smartcreatorhub.blog]. There is not enough verified public evidence to describe the underlying stack, reliability layer, or orchestration logic in detail, and no tech-stack inference is supportable from job postings because no public hiring record surfaced in the research.
Partially corroborated -- Based on one company documentation source and partial corroboration from Crunchbase, Dealroom News, and The Neuron.
Market Research
PUBLIC
The market matters now because AI agents are moving from model experimentation toward action-taking workflows, and that shift creates demand for the infrastructure layer that lets software agents find, invoke, and pay for third-party tools at production scale [Dealroom News, October 2026] [The Neuron, September 2026].
The available public record does not support a clean TAM, SAM, or SOM for Monid's category, and that absence is material. There is no cited third-party market report in the source set that sizes an "AI agent tool marketplace" or "payment layer for agent tools" as a distinct segment, so any numerical market framing here has to be treated as analogous rather than direct. The closest public analog in the evidence is the broader API and developer-platform stack, where the economic logic is familiar: reduce integration work, aggregate supply, standardize access, and add billing rails on top of fragmented providers [Crunchbase] [monid.ai].
That makes the strongest demand signal category formation rather than formal market sizing. Monid is described publicly as a single access layer for AI agents to discover, inspect, run, and pay per call across a large catalog of tools and APIs [The Neuron, September 2026] [Crunchbase]. If that framing holds, the company is not only competing inside API aggregation, it is also testing whether agents become a new buyer and operator class for software infrastructure, with different needs around tool discovery, price comparison, prepaid balances, and low-friction invocation [fundup.ai] [monid.ai].
The product-level growth claims, while unevenly sourced, line up with those tailwinds. Dealroom reported that Monid launched in April 2026 with roughly 200 tools and expanded to more than 1,700 endpoints by September 2026 [Dealroom News, October 2026]. Separate public references describe the product as an "OpenRouter for agent tools" and cite catalogs ranging from 1,700 to 2,000-plus endpoints across dozens of providers [The Neuron, September 2026] [superpowerdaily.com] [monid.ai/openrouter-for-agent-tools]. Those figures do not size the market, but they do suggest two useful points: supply fragmentation is real enough to aggregate, and developers appear willing to consume tools through an abstraction layer if pricing and integration overhead improve.
| Market lens | Public signal | Why it matters |
|---|---|---|
| Direct category | No third-party TAM for AI-agent tool marketplaces in the cited source set | The category is still early, so investors should underwrite on workflow adoption and infrastructure need rather than headline market size |
| Analogous market | API and developer-platform infrastructure, inferred from Monid's one-key access and pay-per-call positioning [Crunchbase] [monid.ai] | Suggests the company sits closer to infrastructure aggregation and billing than to end-user SaaS marketplaces |
| Emerging buyer behavior | Agents need to discover, compare, call, and pay for tools through one integration [fundup.ai] [The Neuron, September 2026] | If agents become persistent software operators, orchestration and payments can become a distinct control layer |
The table points to an early but coherent market shape: the category is easier to describe functionally than to size defensibly. For now, the better question is whether agent-native usage becomes frequent enough for a dedicated access and billing layer to matter.
Adjacent markets are easier to identify than the core market itself. One adjacency is API gateway and developer tooling, where value comes from reducing implementation burden and normalizing access across heterogeneous endpoints [Crunchbase] [monid.ai/docs]. Another is payments infrastructure, where prepaid balances and pay-per-use settlement become part of the product rather than a back-office feature [fundup.ai] [Dealroom News, October 2026]. A third is model routing and inference aggregation, implied by the repeated comparison to OpenRouter, except Monid appears to apply that logic to external tools rather than model endpoints [The Neuron, September 2026].
Those adjacencies matter because substitute behavior is already available. Developers can integrate directly with APIs, stitch workflows together in automation products, or rely on broader integration platforms such as Pipedream and Composio, both named in the research as competitors [Dealroom News, October 2026]. Monid's market opportunity therefore depends less on proving that APIs exist, and more on proving that agent operators prefer a unified registry, execution layer, and billing surface over assembling those components themselves.
Regulatory and macro forces are present, even if the cited record does not tie Monid to any single rule set. A pay-per-call marketplace for external tools sits near several policy-sensitive areas: payment flows, usage-based billing disclosures, API terms compliance, and the governance of autonomous software acting on behalf of users or businesses [fundup.ai] [monid.ai/docs]. More broadly, the macro tailwind is cost discipline in AI application building. A usage-based model with no subscriptions or contracts, as described in Monid's documentation, aligns with buyer preference for variable spend while agent economics are still unsettled [monid.ai/docs]. The counterweight is that variable-spend infrastructure can scale quickly in usage before budgets, controls, and trust standards catch up.
The public evidence supports a cautious read. There is a visible need emerging around agent access, orchestration, and payments, but the category remains too new for clean market sizing from independent third-party research in this source set. That usually means underwriting the company against behavior change, not against a polished TAM slide.
Reasoned from indirect evidence -- Section relies on public product descriptions and funding coverage from Dealroom, The Neuron, Crunchbase, and company materials, but no independent third-party market sizing report for this exact category was available in the cited source set.
Competition and Substitutes
MIXED
Monid is positioning itself between API aggregation and agent infrastructure, with a narrower claim than general developer automation platforms and a broader claim than single-provider routing tools [Crunchbase] [The Neuron, September 2026].
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Monid | Marketplace and payment layer for AI agent tools, enabling discovery, comparison, and pay-per-use API access | Seed, about $9.8M disclosed across pre-seed and seed | Single integration and shared balance for a large catalog of agent-callable tools and APIs | [Dealroom News, October 2026] [Crunchbase] [The Neuron, September 2026] |
| OpenRouter | Model and API routing layer for AI applications | Not established in the provided source set | Framing influence is strong enough that third-party coverage described Monid as an "OpenRouter for agent tools" | [The Neuron, September 2026] |
| Pipedream | Developer automation and API workflow platform | Not established in the provided source set | Broader workflow and integration posture rather than a purpose-built agent checkout layer | [Structured facts] |
| Composio | Tooling layer for connecting agents to external apps and actions | Not established in the provided source set | Closer adjacency to agent action execution than generic API routing | [Structured facts] |
| Arcade | Adjacent agent tooling platform | Not established in the provided source set | Presence in the same buyer conversation, though the overlap is not fully substantiated in public reporting here | [Structured facts] |
The field breaks into three visible groups. First are routing and access-layer products, where OpenRouter is the clearest reference point because public writeups used it as shorthand for Monid's product concept [The Neuron, September 2026].
Monid's edge today appears to be packaging, not a proven moat. Public sources consistently describe one connection, one balance, and pay-per-call access across a large catalog, with catalog counts ranging from more than 1,700 endpoints to roughly 2,000 endpoints depending on source and date [Dealroom News, October 2026] [The Neuron, September 2026] [Crunchbase]. That matters because the developer pain point is fragmentation: separate onboarding, pricing, and billing for each tool. The harder question is durability. If the advantage rests mainly on aggregation and checkout, it is perishable unless transaction volume creates proprietary ranking, pricing, or reliability data that competitors cannot copy. Public reporting confirms more than 4 million agent transactions, but it does not yet show whether those transactions translate into exclusive supply, preferred economics, or meaningful switching costs [Shengkun Ye LinkedIn, retrieved 2026] [en.wowtale.net, September 2026].
The company is most exposed where a competitor owns an adjacent layer that can compress the marketplace function. OpenRouter has the strongest named conceptual advantage because developer mindshare around a single access layer is already associated with its model-routing product, and Monid is being explained through that analogy rather than on a standalone category definition [The Neuron, September 2026]. Composio is the sharper category threat if the market settles on agent action reliability and app-specific depth over breadth, since Monid's public positioning emphasizes catalog scale and billing simplicity more than vertical depth in any one app family [fundup.ai] [monid.ai].
The most plausible 18-month scenario is a split market. Winner if breadth and billing abstraction become the default purchase criterion: Monid, because its current narrative is built around reducing integration overhead and monetizing tool usage one call at a time [Dealroom News, October 2026] [The Neuron, September 2026]. A second branch is also easy to see. Winner if agent builders prioritize execution-grade app connectors over broad catalogs: Composio.
Reasoned from indirect evidence -- This section combines one independently reported comparison frame for OpenRouter with structured-facts competitor names that are not fully corroborated by the captured public source set.
Opportunity
Open sources
If Monid executes, the prize is not a feature business but a default transaction layer for AI agents buying and using third-party tools, a position that can matter if agent workflows consolidate around a small number of trusted access and payment rails [Dealroom News, October 2026] [The Neuron, September 2026].
The headline opportunity is straightforward. Monid is trying to sit between AI agents and the long tail of paid APIs, so developers integrate once, keep one balance, and let agents discover, compare, run, and pay per call across a large tool catalog [Crunchbase] [The Neuron, September 2026] [monid.ai]. That outcome is still early, but not purely aspirational, because public reporting does show rapid catalog expansion from roughly 200 tools at launch in April 2026 to more than 1,700 endpoints by September 2026, alongside reports of more than 4 million agent transactions processed [Dealroom News, October 2026] [Shengkun Ye LinkedIn] [en.wowtale.net, September 2026]. If those usage claims hold up over time, Monid is not merely aggregating APIs. It is testing whether agent commerce can be centralized into a single checkout and routing layer.
The upside paths are different, but they share the same core bet: fragmentation in agent tooling creates room for an intermediary.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Agent checkout standard | Monid becomes the default pay-per-call layer for developers building autonomous agents, similar to how API gateways became default infrastructure in earlier cloud stacks | Continued growth in endpoint breadth and provider coverage, plus developer adoption of a single integration model [Dealroom News, October 2026] [The Neuron, September 2026] | Public sources describe one connection for discovery, execution, and payments across 1,700 to 2,000+ endpoints and 70+ providers, which is the right product shape for standardization if developers prefer abstraction over direct integrations [The Neuron, September 2026] [superpowerdaily.com] |
| Open distribution wedge | The open-source positioning attracts developers first, and the monetization layer follows through hosted routing, billing, and balances | Adoption of the MIT-licensed "OpenRouter for agent tools" framing as a community default [The Neuron, September 2026] | Open source can compress go-to-market friction in developer infrastructure, and Monid has already been described publicly in those terms rather than only as a closed marketplace [The Neuron, September 2026] |
| Provider monetization network | API providers treat Monid as a demand-generation and payments channel for agent-native traffic, making the marketplace itself more valuable as supply deepens | More providers choose distribution through aggregated agent traffic instead of negotiating direct integrations one by one [fundup.ai] [The Neuron, September 2026] | The product is explicitly framed as both discovery and payment infrastructure, which gives providers a reason to list and price through the platform if agent usage becomes material [fundup.ai] [Crunchbase] |
The compounding mechanism would come from density on both sides of the network. More providers increase coverage, which makes the one-key integration more useful to developers; more developer traffic gives providers a reason to join and optimize pricing for agent-native demand [Crunchbase] [The Neuron, September 2026]. Monid's reported expansion from roughly 200 tools at launch to more than 1,700 endpoints within months suggests the supply side can scale quickly, while the reported 4 million-plus transactions indicate at least some early demand is already moving through the rail [Dealroom News, October 2026] [Shengkun Ye LinkedIn]. The second layer of compounding is economic. A pay-per-use model with no subscriptions or contracts lowers adoption friction early, but if Monid becomes the system of record for agent spend, switching costs can emerge through billing workflows, tool selection history, and application logic built around one integration surface [monid.ai] [The Neuron, September 2026].
The size of the win depends on whether this becomes infrastructure or remains a convenience layer. There is no confirmed market sizing in the public record here, so the cleanest comparable is strategic position rather than TAM math. If Monid became a recognized default access and payment layer for agent tools, the company could plausibly resemble an infrastructure asset valued on transaction volume, developer adoption, and ecosystem control rather than on simple seat growth. With $9.8 million in disclosed funding and a product already framed in public reporting as the "checkout counter for AI agents," the upside case is a platform worth hundreds of millions to low single-digit billions of dollars in enterprise value, scenario, not a forecast, if agent transactions continue compounding and Monid captures a meaningful share of that flow [Dealroom News, October 2026] [trysignalbase.com] [The Neuron, September 2026]. That is a large claim relative to the evidence base, but the category logic is real: transaction layers that become defaults tend to matter more than the point products built on top of them.
Partially corroborated -- Relies on one independent named-publisher source for core opportunity framing, with partial corroboration from Crunchbase and additional public but mixed-quality sources.
Sources
Open sources
[fundup.ai] Monid | https://fundup.ai/recently-funded-startups/company/ffa8310f9b81bbed76e6dedce515c31684942f07d755dce974fd01c8ca63d5ca/monid
[Crunchbase] Monid - Crunchbase Company Profile & Funding | https://www.crunchbase.com/organization/monid
[Dealroom News, October 2026] Monid raises $7.7M seed to be the checkout counter for AI agents | https://dealroom.co/news/160477-monid-raises-7-7m-seed-to-be-the-checkout-counter-for-ai-agents/
[chatgate.ai] Monid | https://chatgate.ai/post/monid-2
[trysignalbase.com, September 2026] Monid raises $2.1M pre-seed for agent tool platform | https://www.trysignalbase.com/news/funding/monid-raises-2-1m-pre-seed-for-agent-tool-platform
[The Neuron, September 2026] Everything That Happened in AI Today (Wednesday, September 16, 2026) | https://www.theneuron.ai/digest/everything-that-happened-in-ai-today-wednesday-september-16-2026/
[monid.ai] Welcome - Monid | https://monid.ai/docs
[ZoomInfo] Shengkun Ye | https://www.zoominfo.com/p/Shengkun-Ye/-2082284475
[rocketreach.co] Shengkun Ye email address & phone number | https://rocketreach.co/shengkun-ye-email_60457668
[f.inc] Feiyou Guo | https://f.inc/founders/feiyou-guo/
[Shengkun Ye LinkedIn] Shengkun Ye | https://www.linkedin.com/in/shengkun-ye/
[en.wowtale.net, September 2026] Monid raises US$7.7 million to connect AI agents with pay-per-use tools | https://en.wowtale.net/2026/09/monid-raises-us77-million-to-connect-ai-agents-with-pay-per-use-tools/
[smartcreatorhub.blog] Monid | https://smartcreatorhub.blog/monid/
[superpowerdaily.com] Monid | https://www.superpowerdaily.com/p/monid
[monid.ai/openrouter-for-agent-tools] OpenRouter for Agent Tools - Monid | https://monid.ai/openrouter-for-agent-tools
Articles about Monid
- A $9.8 Million Seed Backs the Checkout Counter for AI Agents — The San Francisco startup has raised $9.8 million to build a unified payment and discovery layer for autonomous software.