LLMCost
Inference provider

Azure: every model it hosts

19 endpoints tracked. Hosts may use different quantizations, context limits, and serving defaults, so compare price together with reliability and configuration. On this host family, the same model costs up to 55× more depending on which endpoint serves it.

Verified 2026-10-21 · source: OpenRouter public API

Cheapest here
OpenAI: GPT-6 Luna
$0.100 in / $0.500 out per 1M
Priciest here
Anthropic: Claude Opus 5
$5.50 in / $27.50 out per 1M
ModelInput /1MOutput /1MCache readContextUptime 1dThroughput
OpenAI: GPT-6 Lunaunknown$0.100$0.500$0.0101050K100.0%—
Anthropic: Claude Haiku 5.5unknown$0.100$0.500$0.0101000K100.0%—
OpenAI: GPT-6 Lunaunknown$0.110$0.550$0.0111050K100.0%—
OpenAI: GPT-6 Lunaunknown$0.110$0.550$0.0111050K100.0%—
Anthropic: Claude Haiku 5.5unknown$0.110$0.550$0.0111000K100.0%—
OpenAI: GPT-5.6 Lunaunknown$0.200$1.20$0.0201050K98.7%—
DeepSeek: DeepSeek V4 Flash 0423unknown$0.210$0.560$0.0311049K75.7%—
OpenAI: GPT-5.6 Lunaunknown$0.220$1.32$0.0221050K100.0%—
OpenAI: GPT-5.6 Lunaunknown$0.220$1.32$0.0221050K100.0%—
DeepSeek: DeepSeek V4 Pro 0423unknown$1.91$3.83$0.1601049K99.9%—
Anthropic: Claude Sonnet 5.5unknown$2.00$10.00$0.1001000K100.0%—
OpenAI: GPT-6.1 Solunknown$2.00$10.00$0.1001050K93.6%—
Anthropic: Claude Sonnet 5.5unknown$2.20$11.00$0.1101000K100.0%—
OpenAI: GPT-6.1 Solunknown$2.20$11.00$0.1101050K100.0%—
OpenAI: GPT-6.1 Solunknown$2.20$11.00$0.1101050K100.0%—
Anthropic: Claude Opus 5.5unknown$4.00$20.00$0.2001000K100.0%—
Anthropic: Claude Opus 5.5unknown$4.40$22.00$0.2201000K——
Anthropic: Claude Opus 5unknown$5.00$25.00$0.5001000K100.0%—
Anthropic: Claude Opus 5unknown$5.50$27.50$0.5501000K——

Prices are list prices at build time; providers run promotions. Uptime/throughput are OpenRouter's rolling measurements. Compare hosts on the providers index.