flexway
FRENTalk to an expert

Market expectations

What enterprises expect from a gateway, and where Flexway stands.

Every expectation of today and tomorrow, each with Flexway's status in the published release; what is already there for a partial one, when unfolded.

Family
Status

Upcoming releases are not commitments; "Delivered" only when the catalogue of the published release says so.

Today 44 expectations

What enterprises expectFlexway v1.6.3
Centralised, secure access to models
Centralise access to providers and internal modelsDelivered · since v0.1.0
Hide provider keys, manage secretsDelivered · since v0.1.0
Authenticate applications, users and services (OIDC, JWT, mTLS)Delivered · since v0.1.0
Rights per team, application, model or environment
Already there, and what is missing

Scopes per endpoint, overrides per tenant and per agent, models allowed per backend; models allowed per caller (tenant, plan, use case, client) are missing

Partly · next in 1.9
Forbid direct access to providers
Already there, and what is missing

Only the gateway holds the keys; forbidding the direct network route is the job of the customer's network egress rules

Partly · next to be scheduled
Routing between models
Choose the model by task
Already there, and what is missing

Routing by capability (chat, vision, embeddings); model aliases and use cases framed; routes per declared task are missing

Partly · next in 1.9
Fail over when a provider is downDelivered · since v0.1.0
Choose by latency, cost, quality or region
Already there, and what is missing

Cost: a soft cap that switches; measured latency, declared cost, region and hosting class as declared criteria are missing. Quality cannot be measured by the gateway (evaluation: out of scope)

Partly · next in 1.9
Send simple requests to a small model
Already there, and what is missing

Possible through a declared route (the task says so); a classifier on the path is excluded by a product rule

Partly · next in 1.9
Tell cloud, private and local models apart
Already there, and what is missing

The external attribute of backends and pinning of sensitive requests; the declared hosting class (cloud, private, local) is missing

Partly · next in 1.9
Cost control
Quotas per application, team, user or project
Already there, and what is missing

Tenant and agent built; user and use case framed; application, team, project and cost centre are missing

Partly · next in 1.8
Limit requests and tokens
Already there, and what is missing

Requests per second per caller, tokens per day and per month; a token rate per minute against peaks is missing

Partly · next in 1.8
Monthly budgets and overrun alerts
Already there, and what is missing

Monthly budget in euros per tenant, gauges and status in the console; budget alert thresholds and built-in rules are missing

Partly · next in 1.8
Control peaks, prevent abuseDelivered · since v0.1.0
Refuse, queue or reroute on overrun
Already there, and what is missing

429 refusal and the tenant's soft-cap switch; graduated actions for each budget level are missing

Partly · next in 1.8
Internal chargeback per cost centre
Already there, and what is missing

Consumption per tenant, CSV export from the console; attribution per cost centre and its export are missing

Partly · next in 1.8
Data protection with models
Detect or mask personal data
Already there, and what is missing

Masking of REST response fields built (redact); in prompts, nothing yet: deterministic detectors and masking

Planned in 1.9
Prevent confidential data from reaching some providers
Already there, and what is missing

A request marked sensitive is pinned to an internal backend; data classes and the backends allowed per class are missing

Partly · next in 1.9
Data residency
Already there, and what is missing

Announced "by policy", not built: regions allowed per tenant or per class, proven in the trace

Planned in 1.9
Filter prompts and responses
Already there, and what is missing

Input and output guardrails in the catalogue from the start, never framed

Planned in 1.9
Detect injection attempts
Already there, and what is missing

Declared patterns and the customer's guard service; the gateway is not a sufficient guarantee (the analysis says so too)

Planned in 1.9
Trace filtering decisions and refusals
Already there, and what is missing

Chained audit of authorisation decisions, closed refusal codes; the guard decision line, without content, is missing

Partly · next in 1.9
AI observability and audit
Latency, availability, errors, tokens, estimated costsDelivered · since v0.1.0
Link each call to an application, user, use case
Already there, and what is missing

Tenant, agent, client; hashed subject in the audit; use case framed; application and cost centre

Partly · next in 1.8
Correlate traces with OpenTelemetryDelivered · since v0.1.0
Analyse models used and error rates
Already there, and what is missing

Per backend and capability; the model served is in the audit, not in the metrics

Partly · next in 1.8
Audit evidence and compliance reports
Already there, and what is missing

Chained audit; exportable execution proof planned; DORA and NIS2 reports on the horizon

Partly · next in 1.9
Follow cost and quality after a model change
Already there, and what is missing

Cost per backend, reload marks, comparison of two windows; quality belongs to an evaluation platform

Partly · next to be scheduled
Agent tool governance
Control agents' tool callsDelivered · since v0.1.0
Expose business APIs as MCP toolsDelivered · since v0.1.0
Authorise by identity and context
Already there, and what is missing

Scopes, roles, unattended agents, delegation chain; contextual control of arguments (an amount by identity) is missing

Partly · next in 1.10
Limit each agent's toolsDelivered · since v0.1.0
Trace agent -> tool -> API -> systemDelivered · since v0.1.0
Require human approval for a sensitive action
Already there, and what is missing

The gateway requires and verifies an approval proof issued by the identity provider; the workflow stays with the customer

Planned in 1.10
Enterprise use cases
Turn existing APIs into AI toolsDelivered · since v0.1.0
Control autonomous agents (look up, not pay)
Already there, and what is missing

Scopes per endpoint, unattended: false; approval and task rights are missing

Partly · next in 1.10
Govern RAG access (search, vectors, embeddings)
Already there, and what is missing

Embeddings under budget, search APIs and vector databases served as endpoints; filtering rights per document is out of scope

Partly · next to be scheduled
Industrialise the prompt lifecycle
Already there, and what is missing

Configuration versioned in Git, models allowed per backend; models approved per caller planned; prompt registry and result comparison: out of scope

Partly · next in 1.9
Cut latency and costs (cache, small model)
Already there, and what is missing

REST response cache; semantic cache planned; routes per task planned

Partly · next in 1.9
Cloud and self-hosted side by side
Already there, and what is missing

Sensitive requests pinned to an internal model, relay to another Flexway; data classes and hosting planned

Partly · next in 1.9
Agent identities, limited rights, validity period
Already there, and what is missing

Declared agent client, delegated token bounded in duration and depth, an exchange that only ever narrows; rights limited to one task are missing

Partly · next in 1.10
Govern AI in IDEs and CI/CD
Already there, and what is missing

Every client goes through the same identity, quota and audit; attribution per application (the client) with its budget is missing

Partly · next in 1.8
Catalogue of AI services
Already there, and what is missing

Catalogue of endpoints, Agents screen (tools, federated servers), declared backends; a unified catalogue (models, agents, tools, owners, terms) is missing

Partly · next in 1.10
Regulatory rules by sensitivity
Already there, and what is missing

Pinning, audit without prompts; models, data and logging per class planned; execution proof planned

Partly · next in 1.9

Tomorrow, by 2030 14 expectations

What they will expectFlexway v1.6.3
Emerging uses
Semantic routing, selection by task, sensitivity, budget, latency
Already there, and what is missing

Planned through declared rules; autonomous selection by a model is excluded

Partly · next in 1.9
MCP gateway and agent governance
Already there, and what is missing

Approved tools = contract, scopes, quotas, inter-agent traces built; contextual arguments, budget per run, registry to come

Partly · next in 1.10
Dynamic authorisation, trusted delegation
Already there, and what is missing

Bounded RFC 8693 delegation built; task rights and approval proof planned, together with the IAM

Partly · next in 1.10
Automatic optimisation of cost, quality, latency
Already there, and what is missing

Recommendations as reviewed proposals (AI for operations); autonomous application is excluded: the console never mutates

Partly · next in 1.10
Model, prompt or agent versions: canary, rollback
Already there, and what is missing

Versions of an upstream; canary between inference backends; prompt versions: out of scope

Partly · next in 1.9
Federation of gateways across entities
Already there, and what is missing

Multi-gateway console framed with no milestone; common policies distributed through GitOps; shared common fragments: candidate

Partly · next to be scheduled
Trajectory 2026-2030
2026-2027 "industrialise": authentication, keys, token quotas, multiple models, costs, observability, first MCP uses
Already there, and what is missing

Foundation built; cost attribution and declared routing planned

Partly · next in 1.8
2027-2028 "govern": tools, agent identities, contextual authorisation, delegation, audit, versions
Already there, and what is missing

Tools, delegation, A2A built; agent actions planned

Partly · next in 1.9
2028-2030 "optimise": federated policies, dynamic selection, arbitration of cost, performance, risk
Already there, and what is missing

Federation: candidate; arbitration by declared rules only

To be scheduled
Predictions
The gateway arbitrates agents' actions
Already there, and what is missing

The contract's endpoint is the business action; approval and task rights planned

Partly · next in 1.10
Edge, cloud and device cooperation; central policies applied locally
Already there, and what is missing

The binary runs locally with no central plane; signed configuration pulled by a site: planned

Planned in 1.11
Decisions tied to the real state of the system
Already there, and what is missing

external-call covers HTTP; asynchronous commands: candidate

To be scheduled
Trusted intermediary between organisations
Already there, and what is missing

Multiple issuers, partner mTLS, token exchange, per-field encryption; declared purpose of use: candidate

Partly · next to be scheduled
"Immune system": revoke, suspend, block a version, isolate, distribute a rule, protect locally
Already there, and what is missing

Revocation by reload, closed by default, egress guards, declared automatic rollback, local protection by design; an automatic suspension decided by a detector is excluded (proposals only)

Partly · next to be scheduled

What Flexway does not do, and who it leaves it to

What the market analysis and Flexway place outside the gateway.

ResponsibilityLeft toWhat Flexway does at the border
Authenticate users and servicesIAM, identity providerConsumes and verifies tokens; never issues any
Agent logic and stateAgent runtime, orchestratorGoverns A2A without terminating it
Document search, per-document rightsRAG platformGoverns access to search and embedding APIs
Evaluate answer qualityLLM evaluation, LLMOpsExposes costs, latencies, errors per backend for that evaluation
Prompts, models, versionsLLMOps registry and pipelineEnforces the declared approved list and the canary decided elsewhere
Human approval of a sensitive operationBusiness workflowRequires and verifies the approval proof
Analyse traces and incidentsOpenTelemetry platformEmits OTLP; the console covers what the gateway sees
Secrets and machine identitiesVault, IAM, PKISecret references, the customer's KMS, mTLS; SPIFFE on the horizon

Let's talk about your gateway

Flexwiz is a young company: you talk directly with the person who founds it. Tell us what you want to run through Flexway. We talk about your infrastructure and your digital sovereignty requirements.

Online booking is coming soon; until then, an e-mail is enough.