We operate the hardware ourselves, giving you direct pricing, predictable performance, and full visibility into the model you are running.
Start with 100,000 free tokens. No card required.
Aggregate throughput at 64 concurrent requests. Requests originated from the same independent cloud environment. Higher throughput and lower error rates are better.
“There were no latency issues using Qwen.”
Agentic coding
A developer tested Qwen on a Rust DSP mastering tool containing a deliberately misleading hardcoded ID. The model attempted the obvious fix, saw its tests fail, backtracked, identified the underlying cause, and produced a patch that two other models independently reviewed.
“Much better than the OpenRouter options we tested.”
Production spam-call classification
After benchmarking earthruntime against OpenRouter-hosted alternatives in production, the team made earthruntime its primary provider and retained the marketplace as an outage fallback.
curl https://api.earthruntime.com/v1/chat/completions \
-H "Authorization: Bearer $EARTHRUNTIME_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen3.6-35b",
"messages": [{"role": "user", "content": "Explain why tail latency matters for production inference in three sentences."}]
}'
curl https://api.earthruntime.com/v1/chat/completions -H "Authorization: Bearer $EARTHRUNTIME_KEY" -H "Content-Type: application/json" -d '{"model":"qwen3.6-35b","messages":[{"role":"user","content":"Explain why tail latency matters for production inference in three sentences."}]}'
Works with your existing OpenAI client
Change the base URL and API key. Your request and response formats stay the same.
Qwen 3.6 served in native BF16
262K context. No undisclosed quantization. Reasoning is disabled by default for lower latency and can be enabled when the workload benefits from it.
No subscription or account setup required.
We operate inference hardware in cities, then look for local uses for its physical outputs. Our first system pairs compute with direct-air carbon capture and supplies recovered CO2 to nearby hospitality businesses.
Faster inference is the product today. A more useful relationship between cities and compute is what we are building toward.
We love to bottle up the CO2 for soda to sell to all our favorite local bars and restaurants ;)