Start Free

Seven Frameworks.
One AI Control Plane.

Framework Ecosystem

Seven Frameworks.One AI Control Plane.

CARDIAC-PURR AI Control Plane integrates with seven agent frameworks and routes each request through the same capability-aware model qualification and tier selection used across the platform. Agent workloads route through the same governed, capability-aware layer as every other request — no separate integration tier, no separate rules. This page covers the architecture and integration paths — for the full AI Control Plane provider benchmark, see Benchmarks.

Integrated

LangChain

Routed through the standard CARDIAC-PURR AI Control Plane model interface.

ReAct · FunctionTool

LlamaIndex

Integrated via its ReAct and FunctionTool agent interfaces — the most specifically-documented integration path of the seven.

Integrated

AutoGen

Multi-agent conversations routed per request through the same capability-aware layer.

Integrated

CrewAI

Crew and task-based agent workflows route through the control plane per step.

Integrated

Haystack

Pipeline-based agent components route through the same model qualification layer.

Integrated

DSPy

Programmatic prompting pipelines route through CARDIAC-PURR AI Control Plane's tier selection.

Integrated

Semantic Kernel

Plugin and planner-based agent execution routes through the control plane.

OpenAI-compatible

Compatible integration path

An OpenAI-compatible endpoint is available for applicable frameworks — point your existing client at the CARDIAC-PURR AI Control Plane base URL without a framework-specific adapter.

Architecture

The AgenticControl Plane.

Agent Routing & Capabilities.

Agent Framework Layer

LangChain
LlamaIndex
AutoGen
CrewAI
Haystack
DSPy
Semantic Kernel

CARDIAC-PURR AI Control Plane

Request requirements → Capability qualification → Complexity / tier routing

Tool calling
Structured output
Reasoning
Context window
Streaming
Vision

Provider + Model Selection

Anthropic
OpenAI
Google
Mistral
Cohere
DeepSeek
Grok
Azure OpenAI
Returned to the originating agent framework

These six dimensions are tracked as capability metadata within the routing layer. A per-provider capability matrix with specific values is not yet published — the diagram above reflects the qualification stages that exist, not a scored comparison.

Routing

Automatic.Or Explicit.

# Let the router decide the tier router.query(prompt, model="auto") # Or choose explicitly router.query(prompt, model="small") router.query(prompt, model="medium") router.query(prompt, model="large") # Or force a tier for this call only router.query(prompt, force_tier="large")
Current Platform

Live tier-resolution trace

A current v114 validation call using explicit LARGE-tier routing:

model="large"tier=largeclaude-opus-4-8cache_hit=false

Reflects current production configuration (build cardiac-purr:v114). This is a capability trace, not a benchmark result, and is separate from the dated benchmark evidence below.

What "auto" does today

model="auto" evaluates the request and selects a tier without manual specification. In the published agentic benchmark below, this resulted in every one of the 350 factual-QA requests routing to the small tier — consistent with a workload that didn't require medium- or large-tier capability.

Recorded Benchmark Evidence

One Dated Run.Clearly Scoped.

Run August 12, 2026
Scope 7 frameworks · 50 requests each · 350 total
Workload Factual‑QA
Model Anthropic · claude‑haiku‑4‑5‑20251001 · small tier
100%
Accuracy on this factual-QA benchmark workload
80%
Savings on this benchmark workload

What this benchmark does & doesn’t show

  1. 01These figures apply only to the published 350-request factual-QA workload — they are not a general accuracy or savings claim.
  2. 02Every request in this benchmark routed to the small tier; medium and large tiers were not exercised by this run.
  3. 03Caching affected the workload. This benchmark reflects a cached-influenced result, not a purely cold-inference measurement.
  4. 04Results are reported as an aggregate across all 7 frameworks combined — no individual per-framework breakdown is published from this run.

Evidence Boundaries

What This PageDoes Not Claim.

Structured-output, reasoning, context-window, and vision awareness are tracked as capability metadata dimensions within the routing layer. No benchmark results are currently published for these specific dimensions — only the factual-QA workload above has been benchmarked.

Streaming is supported in current integration code. No streaming-specific benchmark results are published on this page.

No per-provider capability matrix with scored values is published. The architecture diagram above reflects qualification stages that exist in the routing layer, not a comparative scorecard.

No individual framework (LangChain, CrewAI, etc.) has a published standalone accuracy or savings figure — only the 7-framework aggregate from the August 12, 2026 benchmark.

FAQ

CommonQuestions.

Which agent frameworks does CARDIAC-PURR AI Control Plane integrate with?

Seven frameworks: LangChain, LlamaIndex, AutoGen, CrewAI, Haystack, DSPy, and Semantic Kernel. Integration mechanics differ by framework — LlamaIndex, for example, is integrated via its ReAct and FunctionTool agent interfaces. An OpenAI-compatible endpoint is also available for applicable frameworks.

Does the router pick the model automatically, or can I choose?

Both. model="auto" lets the router select a tier automatically. You can also force an explicit tier with model="small", model="medium", or model="large", or override per-call with force_tier in router.query().

What does the published agentic benchmark actually measure?

A 350-request factual-QA workload (50 requests across each of the 7 frameworks), run August 12, 2026, entirely on the small tier via Claude Haiku 4.5, with caching present. It does not measure medium- or large-tier behaviour, reasoning-heavy tasks, or tool-use-heavy workloads.

Is this the same benchmark as the AI Control Plane benchmarks page?

No. The AI Control Plane benchmark measures routing accuracy and cost savings across 9 providers on general enterprise queries, run July 12, 2026. The agentic benchmark on this page is a separate, framework-specific study run on a different date against a different workload.

What's the difference between the benchmark model and the current platform's large-tier model?

The August 12, 2026 benchmark used claude-haiku-4-5-20251001 at the small tier. The current production platform's large tier routes to claude-opus-4-8. These are separate and should not be compared as if they were the same test.

Evaluate Agentic Routing on Your Own Workflows.

Technical documentation and a scoped evaluation plan are provided at the start of every engagement.

Validate on your own traffic first Commercial terms confirmed in writing Claims reviewed before publication