Model library

Build with leading open models

Browse top-performing models across chat, vision, image, audio, embeddings, and rerank — unified behind stable HTTP contracts and tenant-safe isolation.

Already have access?Create an account
ML
LogReg logo
LogReg
XGBoost logo
XGBoost
GenAI providers
Qwen logo
Qwen
Meta logo
Meta
OpenAI logo
OpenAI
BFL logo
BFL
DeepSeek logo
DeepSeek
XALORRA RUNTIME
Model Library
Unified routing across providers with stable contracts, tenant isolation, and reproducible runs.
DeepSeek R1
Qwen 2.5 Coder
Llama Vision
FLUX.1 [pro]
Embeddings
Rerank v1
Featured models

Fast picks for production deployments

These are curated defaults: strong reasoning, reliable coding, and high throughput. Everything stays tenant-safe and auditable.

DeepSeek logo
DeepSeek R1
DeepSeek
New
High-reasoning model for complex analysis and coding. Strong for tool-augmented workflows.
Reasoning
Coding
Long context
Capabilities: chat, code
Try this model
MO
Kimi K2 Instruct
Moonshot
Fast
Agent-ready instruct model optimized for throughput and reliable completions.
Fast
Agentic
General
Capabilities: chat, code
Try this model
Qwen logo
Qwen 2.5 Coder
Qwen
New
Code-focused model for generation, refactors, and patch-style reasoning.
Code
Refactor
Tools
Capabilities: chat, code
Try this model
Tip: use the filter pills below to browse by capability.
Target: /studio/d7c78de1-ef75-44ee-8e89-37ee95e2cc89

Browse models by capability

Filter by workload type, search by provider/family, then wire it into agents and dataset-scoped tools.

Results: 9
Deployment
Swap providers without rewriting your product.
Xalorra keeps your routing, tenant isolation, dataset-scoped tools, and auditability consistent. Your app stays stable while you iterate on models.
Inference

Run any model on the fastest endpoints

Use a single API to route open models with stable contracts and predictable behavior. Keep your product consistent while you swap providers behind the scenes.

RUN INFERENCE