Blog
Technical articles on inference optimization and building with LLMs.
Technical report
Quantifying LLM Cost Reduction with Infravelle's Cache-Aware Inference Routing
Statistical report on cache-aware LLM inference arbitrage: methodology, robustness checks, and aggregate results across providers.
Read the reportInfravelle vs OpenRouter: which LLM gateway should you use?
A practical comparison of OpenRouter and Infravelle across model coverage, provider routing, cache-aware cost modeling, pricing, and BYOK.
Read postUse Infravelle with Hermes Agent, OpenClaw, and Kilo Code
Infravelle works as a provider in all three. One API key, models from Anthropic, DeepSeek, Google, and others.
Read postUse the Response API with DeepSeek, Gemini, Claude, and every model on Infravelle
The Response API format works across all 200+ models on Infravelle, with streaming, tool calling, and multi-model routing.
Read postUse DeepSeek, Kimi, GLM, and other open-source models in Claude Code with Infravelle
Point Claude Code at Infravelle to access open-source and open-weight models through one API key.
Read postUse DeepSeek, Kimi, GLM and 200+ models in OpenCode with Infravelle
Infravelle is a registered provider in OpenCode. One environment variable, 15 models out of the box, 200+ with a config line.
Read postThe variables that affect your inference cost
How provider caching mechanics interact with your workload to determine inference cost, and how to optimize for it.
Read post