Requesty

Name: Requesty
Brand: Requesty
Rating: 5 (3 reviews)

AI Native

Requesty is an OpenAI-compatible API gateway and router that proxies requests to 600+ LLM models/providers, providing intelligent routing and failover, prompt caching for token-cost reduction, governance controls, and observability metrics.

Est 2023|Valuation $11.4M|Raised $3.6M|23 Ref Customers

LinkedIn 3K|Website|FTEs 5|HQ London, United Kingdom

Published Oct 15, 2026

01Common complaints

- Company sizes or stages where this product is overkill or insufficient - Very early prototypes that rely on a single model/provider: adding a gateway may increase complexity without clear ROI. - Very large regulated enterprises that require strict network isolation and VPC-only deployment: a self-hosted gateway is likely more appropriate. - Industries or use cases where competitors are clearly superior - Highly regulated or air-gapped environments: self-hosted solutions (e.g., LiteLLM) are typically preferable. - Technical requirements that this product cannot meet - Use cases that require exact parity with every upstream provider’s API quirks; any OpenAI-compatible abstraction can encounter edge-case incompatibilities. - Budget constraints that make alternatives more appropriate - If a percentage-based gateway fee is unacceptable at projected scale, evaluate self-hosted alternatives or negotiate enterprise terms with vendors. - Specific workflows or integrations where the product falls short - Community reports indicate tool-specific integration issues (e.g., Cline) and request validation constraints (e.g., RooCode/cache_control block limits). - Red flags in the buying process - Limited independent reviews and analyst coverage relative to larger competitors. - Confusing or mismatched third-party listings (e.g., a G2 profile that describes different product functionality), which can complicate due diligence. - Current limitations acknowledged by the vendor - Not explicitly listed in the vendor excerpt provided for this research. - Features on the roadmap (not yet available) - Roadmap items are not systematically documented in the accessible public sources captured in this research. - Scale limitations (at what volume/size does it break down?) - The vendor reports high daily token volumes; independent load and performance validation was not available in the accessible sources. - Security or compliance gaps - SOC 2 Type II is claimed; purchasers should review the SOC 2 report scope, subprocessors, and the vendor’s process for handling upstream provider interactions. - HIPAA readiness claims should be verified with documentation and Business Associate Agreement (BAA) availability and scope. - Technical debt or architectural limitations mentioned in reviews - No substantial commentary on vendor technical debt was found in credible public sources for this research. - API limitations or integration constraints - Community reports noted specific integration problems (e.g., Cline not working for some users) and RooCode-related 400 errors tied to cache_control block limits, indicating that client behavior can surface API constraints.

Why some customers left for competitors (churn reasons)

No direct statements of customers switching away from Requesty were found in the captured sources.
Common churn drivers in this category (and implied in community comments) include fee sensitivity, integration reliability with specific clients, and a preference for self-hosting to retain keys/data in-house.

02Why they buy

- Competitive advantages validated by customer feedback (not just vendor claims) - Limited independent validation is available in public sources. Community discussions indicate that Requesty is evaluated as an OpenRouter alternative in some developer/tooling contexts. - Unique features and capabilities competitors lack - No independently verified exclusive feature was identified in public sources. The combination of a stated flat markup, EU endpoint/data residency option, and an integrated bundle (governance, caching, routing, dashboards) is positioned by the vendor as a differentiator. - Differentiation from competitors based on reviews - Credible, high-volume reviews specific to the LLM gateway offering are limited in accessible public sources. - Value propositions with evidence from case studies - Vendor-published case studies contain outcome claims; independent verification was not found in accessible sources. - Areas where competitors clearly have advantages - OpenRouter: broader ecosystem visibility and documented fee thresholds in public materials. - LiteLLM: self-hosting control that avoids per-token gateway fees. - Specialized observability stacks: deeper experimentation workflows and prompt/versioning features in some pairings of proxies and observability tools.

Why customers chose this product over alternatives (inferred from discussions)

Convenience of an OpenAI-compatible unified endpoint.
Desire for routing/failover and cost controls without building an in-house gateway.
The availability of an EU endpoint/data residency option may appeal to EU-based teams.

03Field reports

Review platforms (G2)

- Important caveat: The G2 page for Requesty contains copy that describes a customer interaction analytics product (CSAT/NPS insights), which is inconsistent with the LLM gateway/router positioning. This suggests a historical product pivot, a listing mismatch, or an outdated profile. - Positive review sentiment (from the accessible excerpt): a reviewer indicated Requesty helped them identify trends and issues from customer messages and survey responses, which appears to reference analytics functionality rather than an LLM gateway. - Criticism (from the accessible excerpt): the same reviewer noted limitations in drilling into specific comments. - Takeaway: G2 content may not be reliable evidence about the LLM gateway product specifically given the apparent listing content mismatch.

Community/social (Reddit)

- Positive/neutral mentions: - Users discuss Requesty as an API provider and compare it to OpenRouter in developer tooling contexts. - Negative/critical mentions: - Pricing sensitivity: a community user reported that Requesty was comparatively expensive. This is a subjective comment and context-dependent. - Integration friction: a user reported difficulty getting Requesty to work with Cline and noted interoperability issues with RooCode, suggesting compatibility/implementation challenges in some client contexts. - Error/limitations: a RooCode-related post showed an HTTP 400 error related to cache-control blocks (an error indicating a capped number of cache_control blocks), implying edge-case request formatting or provider-compatibility constraints.

Hacker News

A direct "Show HN" thread for Requesty was not found in the accessible results for this research.
Hacker News threads on LLM routers discuss adjacent competitors; Requesty does not appear as a central focus in the accessible items.

Vendor-published case studies

The vendor blog contains multiple case study-style posts (e.g., e-commerce scaling, fintech/KYC) authored by the vendor. Metrics and outcomes in these posts should be treated as vendor-reported and unverified unless corroborated by the customer or an independent third party.

Independent case studies

No independent analyst-authored case studies or third-party validated ROI reports were found in the accessible sources for this research.

Critical evaluation

Vendor claims of very large token cost reductions are plausible in principle via caching and routing but are highly workload-dependent.
Purchasers should request a methodology (baseline assumptions, token mix, cache hit rates), a sample evaluation on their own traffic, and clarity on what data is cached and how PII masking interacts with caching.

04Best fit

Ideal Customer Profile

- Company sizes: specific employee counts and revenue ranges - Likely best fit: - Small-to-mid sized product and platform teams (small to mid-sized organizations) building AI features that require multiple models/providers. - Mid-market and larger enterprise teams that need governance, RBAC, and budget controls, but that prefer a managed gateway to building an in-house solution. - Less suitable: - Very large enterprises with strict data locality and private network requirements that require full on-prem/VPC-only solutions. - Industries and verticals where adoption is strongest (with customer examples) - Vendor blog targets e-commerce and fintech examples; independent verification of named customers and outcomes was not found in the accessible sources. - Roles and personas: decision makers AND end users - Decision makers - Head of Engineering / Platform Engineering - CTO - Security / Compliance (for governance and data residency concerns) - FinOps / Engineering managers (spend controls) - End users - AI/ML engineers - Backend engineers integrating LLM APIs - Developer productivity teams using coding agents and related tools - Pain points and challenges the product addresses - Reducing dependency on a single model provider - Mitigating provider outages via failover - Implementing cost controls via routing, caching, and budgets - Centralizing governance (approved models, RBAC) - Qualification criteria: indicators of a good fit - Organizations already using OpenAI-style APIs that want a minimal integration change (for example, by swapping an API base URL). - Organizations using multiple providers/models that need unified billing, key management, or unified policy enforcement. - Organizations that require an EU endpoint/data residency option and that seek a vendor-stated SOC 2 Type II posture. - Deal disqualifiers: indicators of a poor fit - Organizations that require a fully self-hosted gateway with no SaaS dependency in the request path. - Organizations that cannot accept any additional latency introduced by an external gateway. - Organizations with stable single-provider usage patterns that do not require multi-provider routing.

05The pitch

What the product does (core value proposition in objective terms)

- Requesty is described by the vendor as an "AI Gateway for Production": an OpenAI-compatible API gateway/router that sits between an application and multiple upstream LLM providers/models. - The vendor reports support for hundreds of models/providers behind a single base URL (vendor-hosted API endpoint) and lists features including: - Intelligent routing (cost/latency/availability-based) - Failover chains and load balancing - Prompt caching (vendor reports substantial cost reductions) - Governance (RBAC, model allowlists, PII masking) - Observability (cost, latency, TTFT, percentile latencies, errors, cache savings) - A vendor-hosted EU endpoint and EU data residency options

Target market and market positioning

Marketed toward teams building production LLM applications/agents that require multi-provider optionality without rewriting integrations, centralized policy and spend control, and reliability via failover across providers.
Also targeted to developer tool ecosystems where OpenAI-compatible endpoints are common (vendor lists integrations such as Cursor, Cline, Continue on its site).

Get a comprehensive analysis of Requesty including market position, competitive landscape, adoption trends, and peer benchmarks.

Download full report →