OpenRouter has integrated OpenAI's GPT-6 Astra, a flagship model optimized for complex, long-horizon agentic tasks such as software engineering, scientific research, and browser-based automation. The model features a substantial 1,050,000 token context window and supports up to 128,000 completion tokens, with capabilities for structured JSON outputs and tool calling. OpenRouter provides multiple routing modes—Balanced, Nitro, Floor, and Exacto—to optimize for cost, speed, or tool-calling accuracy across providers including OpenAI, Azure, and Amazon Bedrock. Pricing is structured at $10.00 per million input tokens and $50.00 per million output tokens, with specialized rates for prompt caching ($1.00/M for reads, $12.50/M for writes) and web search ($10.00 per 1K calls). Performance metrics indicate a P50 throughput of 54 tokens per second and a P50 latency of 3.23 seconds. The model demonstrates high reasoning capabilities, achieving a 94.6% score on the GPQA Diamond benchmark. OpenRouter ensures high reliability through automatic failover; recent data shows 99.63% availability and 100% uptime across several key providers over a 3-day period.