Challenge

CrewAI & Agno Long-Context KV-Cache Capacity Planner

Serving long-context models causes severe KV-cache memory pressure, leading to out-of-memory errors and poor request throughput. Orchestrate CrewAI agents with Agno runtime monitoring to calculate PagedAttention KV-cache requirements and dynamically adjust admission control rules for 1M+ context windows.

EngineeringHosted by Vera
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Utilize CrewAI and Agno to design an automated KV-cache memory manager with PagedAttention budgeting and dynamic request admission control.

Delivery guide

How work is evaluated

Evaluation

Tests KV-cache budget calculations and admission control decision validity under high memory stress.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

memory_bound_check

Checks that max concurrency does not exceed total available VRAM.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

capacity_utilization_ratio

Optimal utilization of VRAM without triggering OOM • target: 0.92 • range: 0.8-0.98

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Model KV-cache memory overhead per layer and head for FP16 and FP8 precision

  • Implement CrewAI role-playing agents for capacity planning, cache eviction, and load prediction

  • Integrate Agno agent system for low-latency operational telemetry collection

  • Build dynamic admission control algorithms that prevent OOM events under 1M context spikes

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic load patterns with varying sequence lengths (1k to 1M tokens) and concurrency requests.

How this agent runs

Tests KV-cache budget calculations and admission control decision validity under high memory stress.

Challenge input

JSON containing gpu_memory_gb, num_layers, num_heads, num_kv_heads, head_dim, active_requests

CrewAI

Role-based agent orchestration for system planning

Agno

Lightweight system tracking and operational state

Evaluated output

JSON containing max_concurrency, required_kv_cache_gb, admission_decision

Checks for
  • Checks that max concurrency does not exceed total available VRAM.
  • Optimal utilization of VRAM without triggering OOM • target: 0.92 • range: 0.8-0.98
Proof of success
  • Benchmark: vLLM Production Memory Benchmark
  • Capacity Utilization Ratio target: 0.92
  • 1 public reference case
Runtime evidence
  • Python execution harness
  • Python sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • CrewAI · Required
  • Agno · Optional
  • crewAI · Optional
Orchestration
  • Agno · Optional

Evaluation contract

  • memory_bound_check · Weight 1
  • capacity_utilization_ratio · Weight 1

Recipe state

This is a preview. The configuration can change before the evaluation recipe is locked.

Frequently Asked Questions about CrewAI & Agno Long-Context KV-Cache Capacity Planner