Requesty LLM Gateway
13 model routes currently listed at zero price.
https://router.requesty.ai/v1Models mentioned
nvidia/muse-glimmer-30bnvidia/nemotron-3.5-content-safetynvidia/nemotron-3-super-120b-a12bnvidia/nemotron-3-ultra-550b-a55bnvidia/nemotron-3.5-lightning-30b-a3bnvidia/nemotron-3-nano-30b-a3bnvidia/nemotron-3-nano-omni-30b-a3b-reasoningmistral/leanstral-1-5novita/inclusionai/ling-3.0-tinynovita/ling-3.1-flashgoogle/gemma-4-31b-itpoolside/laguna-m.1poolside/laguna-xs.2Equivalent paid value
How this was valued: Maximum among current zero-price routes with a verified same-base-model paid comparator, for one shared 200-requests/day pool. Nemotron Ultra advertises 1,048,576 total-context tokens and 65,536 max output; include the remaining 983,040 input tokens instead of the previous output-only calculation. Prefer Requesty own current standard paid routes: Doubleword $.50/M input + $2.50/M output has the lowest cost for this identical token workload among current Ultra standard candidates (Fireworks NVFP4 $.60/$2.40 and Nebius FP4 $1/$3); discounted :flex is a different service class and excluded. Comparisons are same base model, not identical upstream hosting, quantization, context, tools, privacy or SLA; a smaller paid context can require multiple paid calls. Muse Glimmer operator prefixes are resolved by explicit model_lab/meta and equal canonical_name, not fuzzy model substitution. Catalog rows are shared-pool alternatives, never additive. Other models without exact prices or usable caps remain unpriced. Assumes uninterrupted saturation without latency, capacity, concurrency, availability, or fair-use loss; no authenticated free-route generation was tested.
Limits and terms
- Requests Per Day
- 200
- Reset
- daily
- Caveat
- The Free plan is restricted to zero-priced models; PAYG and BYOK catalogs are separate.
What happens to your prompts?
Free self-service traffic is logged by default and can use upstream models that train on content.
- Plan Scope
- Requesty LLM gateway. Self-service and Enterprise logging defaults differ, and each selected model provider has independent data rules.
- Prompt Retention
- Self-service plans store prompts and outputs encrypted in the EU for up to 30 days by default. Disabling logging enables gateway ZDR prospectively; Enterprise logging is disabled by default.
- Response Retention
- Same plan and logging-setting policy as prompts. Encrypted backups can remain up to 30 days after deletion.
- Ordinary Logging
- With content logging disabled, token counts, model ID, and timestamps remain for billing and statutory accounting retention of six years.
- Model Training
- Requesty itself does not train models on customer content. The privacy summary says free-plan accounts can be associated with models whose upstream provider trains on content; these are labeled Training Permitted Models and are free.
- Product Improvement
- Requesty uses aggregated, de-identified usage for performance, capacity planning, reporting, and service improvement rather than training on identifiable prompt content.
- Human Or Operator Access
- Logged content can be accessed for debugging, support, security, and legal obligations under access controls. ZDR prevents Requesty content logging but not upstream provider access.
- Subprocessors And Routing
- Requests go to the selected third-party model provider. Requesty's EU endpoint constrains its own routing and storage geography, not necessarily the provider's inference location or policy.
- Deletion Controls
- Workspace admins can disable logging at any time for new requests. Logged content expires within 30 days, backups within another 30 days, and account deletion is available subject to statutory records.
- Caveat
- ZDR only controls Requesty's layer. Free Training Permitted Models can allow the final provider to retain or train on prompts even when gateway logging is off.
Governing documents
- Terms Of Service
- https://www.requesty.ai/terms
- Privacy Policy
- https://www.requesty.ai/privacy
- Data Processing Agreement
- https://www.requesty.ai/dpa
Eligibility
- Account Required
- Yes
- Payment Method Required
- No
Primary sources
Before you build with Requesty LLM Gateway
Read the classification narrowly
This record has current first-party evidence for some zero-cost hosted inference. The label describes the bounded offer supported by evidence on 2026-10-06; it does not guarantee permanence, production suitability, uptime, latency, model quality, or access from every region.
Resolve the live model route
This snapshot records 13 model IDs. Match the exact ID against the provider's live catalog before using it in code; zero-price routes and aliases can rotate while an older documentation page remains online. The recorded base endpoint is https://router.requesty.ai/v1, but the provider's current API reference remains authoritative for paths, authentication, and request shape.
Confirm account and billing boundaries
An account is required, so public catalog evidence cannot by itself prove the limits, balance, or billing behavior attached to your workspace. The reviewed evidence explicitly says no payment card is required. Re-check the signup flow because eligibility and billing controls can change after the snapshot. The equivalent paid value shown here prices a documented allowance where possible; it is not cash, guaranteed savings, or protection from overage.
Re-read the prompt policy for your plan
This record classifies the reviewed handling as “Not private”: Free self-service traffic is logged by default and can use upstream models that train on content. Provider policies can distinguish free, paid, enterprise, opted-in, and feature-specific traffic, so verify the governing terms for the exact account and route that will receive your data.
Follow the evidence, then re-check it
The record links 10 first-party sources covering the offer, catalog, limits, pricing, terms, privacy, or adoption evidence available to the audit. Prefer the newest governing document or live catalog when sources disagree, and submit a correction when a provider changes a material term.
Match the quota to the workload shape
Translate the published allowance into the traffic pattern you actually expect instead of comparing headline totals alone. A daily token pool can look generous while a low requests-per-minute or concurrency ceiling blocks interactive bursts; a high request limit can still fail a long-context job when input, output, or per-request tokens are capped. Separate prompt tokens from generated tokens, include retries and tool calls, and test the largest realistic payload. If the provider documents more than one limit window, the tightest window at your peak load is the practical ceiling. Research, experimental, and community access can also carry eligibility or fair-use constraints that cannot be modeled as a simple number.
Plan fallback without changing the rules
A fallback should preserve more than API syntax. Confirm that the substitute route supports the required modality, context length, streaming behavior, structured output, tools, and safety controls, then compare its prompt-retention and training terms. An OpenAI-compatible request shape does not make providers operationally or contractually equivalent. Decide which errors may trigger a retry, cap retry storms, and prevent an exhausted free route from silently switching to a billable model. If deterministic output matters, record the model revision and sampling settings; rotating aliases and free-model pools can change behavior even when the endpoint remains available.
Monitor the offer as a dependency
Capture the model ID, response model field, rate-limit headers, usage fields, latency, HTTP status, and any provider request identifier for each test. Watch for authorization failures, quota exhaustion, catalog removal, policy revisions, and dashboard balance changes as separate failure modes. Re-check the provider’s live catalog and governing pages on a schedule proportionate to the workload’s importance, and keep the dated sources that supported your decision. Free capacity is especially suitable for prototypes, evaluations, fallbacks, and bounded workloads when the application can tolerate change; a production dependency still needs observability, an exit path, and an owner responsible for re-verification.
Test one complete request before scaling
Start with the smallest permitted request using the exact credential, model ID, endpoint, and account type you intend to deploy. Record the HTTP status, response headers, usage fields, latency, and any dashboard balance change. Then exercise the failure path: an invalid model, an exhausted quota, or a rate-limit response should fail clearly without silently switching to a paid route. If the provider supports streaming or tool calls, validate those features separately because a free model can expose a narrower capability set than its paid counterpart. Keep a budget ceiling outside the application whenever billing is possible, and avoid sending sensitive data until the observed route matches the reviewed data agreement.