Domain-Specific Code Models

Each model is tuned for infrastructure or code-heavy workflows. Pull the weights for local use, or chat in-browser using our hosted inference.

Q
Qwen K8s Assistant – 0.5B
qwen_0.5b_k8s-qlora_v4
Kubernetes Params: 0.5B

Ultra-lightweight Kubernetes helper for manifests and object scaffolding, ideal for fast iteration, CI jobs, and resource-constrained environments. Finetuned to outperform base Qwen on core K8s authoring tasks.

Lightweight K8s Manifests
Great for quick checks, small clusters, and CI helpers.
Q
Qwen K8s Specialist – 1.5B
qwen_1.5b_k8s-qlora_v3
Kubernetes Params: 1.5B

Larger Qwen-based Kubernetes model for multi-step reasoning, production workflows, and richer manifest generation. Finetuned to exceed baseline Qwen on troubleshooting, CI/CD, and rollout planning.

Reasoning K8s Ops CI/CD
Use when you want stronger reasoning around cluster-level changes.
L
Llama 3.1 – K8s Expert 8B
llama_v4
Kubernetes Params: 8B

Mid-sized Llama model deeply finetuned for controllers, RBAC, operators, and production-grade deployment patterns. Opinionated best-practice defaults for real-world cluster architectures.

RBAC Controllers Prod-grade
Best when you want both code and higher-level guidance.
G
Gemma 2 – K8s Deep Reasoner 2B
gemma_2b_k8s-qlora_v4
Kubernetes Params: 2B

Gemma-based Kubernetes model tuned for long-form explanations, migration plans, debugging sessions, and architecture reviews. Finetuned to outperform base Gemma on complex K8s design tasks.

Deep reasoning Architecture Debugging
Reach for this when you want narrative “what & why.”