# axionic-docs

## Docs

- [Authenticate the Mechanex SDK with API keys and JWT](https://docs.axioniclabs.ai/products/mechanex/authentication.md): Set up API key and JWT authentication for the Mechanex SDK, persist credentials in the config file, and manage key rotation and token refresh.
- [Mechanex CLI commands for auth, credits, and serving](https://docs.axioniclabs.ai/products/mechanex/cli.md): Reference for every mechanex CLI command including signup, login, API key management, credit balance, top-up, and local server startup.
- [Text generation API reference](https://docs.axioniclabs.ai/products/mechanex/generation.md): Call generation.generate() with ten sampling methods, steering vectors, JSON schema constraints, speculative decoding, and ensemble sampling options.
- [Mechanex Python SDK for inference and steering](https://docs.axioniclabs.ai/products/mechanex/introduction.md): Install and configure the Mechanex Python SDK for remote or local inference, steering vector generation, SAE behavior monitoring, and policy management.
- [Reusable inference policies](https://docs.axioniclabs.ai/products/mechanex/policies.md): Bundle steering, sampling, constraints, and verification into reusable policies you can save, compare, auto-tune, and apply across requests.
- [SAE behavior monitoring and runtime drift correction](https://docs.axioniclabs.ai/products/mechanex/sae.md): Create behavior rules with sae.create_behavior() and generate text with real-time SAE-based drift detection and automatic correction vectors.
- [Local Server (OpenAI-Compatible)](https://docs.axioniclabs.ai/products/mechanex/serving.md): Launch a local OpenAI-compatible server with mx.serve() for drop-in inference, steering vector support, and SAE behavior monitoring on any model.
- [Steering vectors with the Mechanex SDK](https://docs.axioniclabs.ai/products/mechanex/steering.md): Use the Mechanex SDK to compute steering vectors with CAA or few-shot methods, evaluate effectiveness, and save or load vectors for reuse.
- [Behavior rules and drift correction](https://docs.axioniclabs.ai/products/spectra/behaviors.md): Create SAE-based behavior rules that monitor model activations during inference and automatically apply correction vectors when drift is detected.
- [Billing and private model limits](https://docs.axioniclabs.ai/products/spectra/billing.md): Understand how Spectra charges for inference and training usage, how to add credits through Stripe, and the private-model storage limits for each account tier.
- [Fine-tune small language models with SFT and GRPO](https://docs.axioniclabs.ai/products/spectra/finetuning.md): Train a custom model in Spectra using tool-calling or text dataset workflows, supervised fine-tuning (SFT), and optional GRPO reinforcement learning phases.
- [Spectra platform for fine-tuning and steering models](https://docs.axioniclabs.ai/products/spectra/introduction.md): Get started with Spectra, Axionic's web platform for fine-tuning small language models, shaping runtime behavior with steering vectors, and monitoring usage.
- [Browse foundation models and your fine-tuned models](https://docs.axioniclabs.ai/products/spectra/models.md): Browse Axionic-hosted foundation models, review your fine-tuned models, and manage drafts, active training runs, and export integrations in Spectra.
- [Observe API requests, prompts, and SAE monitoring](https://docs.axioniclabs.ai/products/spectra/observe.md): Review API request activity, inspect prompts and responses, and check SAE behavior-monitoring readiness for every model you operate in Spectra.
- [Optimization workspace for steering, sampling, and rules](https://docs.axioniclabs.ai/products/spectra/optimization.md): Configure model selection, steering vectors, behavior rules, sampling, test generations, and API snippets in the Spectra Optimization workspace.
- [Sample Project: Chat Indexer Agent](https://docs.axioniclabs.ai/products/spectra/sample-project.md): Step-by-step guide to building, training, and deploying a fine-tuned 0.5B model for structured data extraction using Spectra distillation.
- [Account and API key settings](https://docs.axioniclabs.ai/products/spectra/settings.md): Manage your API keys, connected OAuth accounts, teacher model keys, HuggingFace integration, and notification preferences in Spectra settings.
- [Define tool schemas for fine-tuned model function calls](https://docs.axioniclabs.ai/products/spectra/tool-schemas.md): Define tool schemas in JSON or natural language so your fine-tuned model learns to call APIs, databases, and external functions accurately.
- [Steering Vectors in Spectra](https://docs.axioniclabs.ai/products/spectra/vectors.md): Learn how steering vectors nudge model behavior at inference time without modifying weights, and manage them inside the Spectra Optimization workspace.

## OpenAPI Specs

- [openapi](https://docs.axioniclabs.ai/api-reference/openapi.json)