Clarke 1.0
Radium's model for everyday production traffic: answers from retrieved documents, copilots, support conversations and structured output.
Clarke is the model Radium puts on most production traffic: answers from retrieved passages, copilots inside the tools your team uses, customer conversations and structured output.
Radium selects open-weight models, tunes them and serves them behind one API, where a router and a reinforcement learning engine choose the model for each request.
Most production traffic goes to Clarke.
RAG
Answers from retrieved passages. Your vector database, embeddings and reranker stay.
Copilots
Assistants inside the tools your team works in, answering from your own docs.
Support
Customer conversations in your voice and on your policies, at support volume.
Code
Tests, docs and explanations, with Hal taking the multi-file changes.
Five times faster to first text than Claude Sonnet.
Clarke and Claude Sonnet ran the same 80 structured tasks with fixed answers, then 40 tasks at four concurrent requests for speed. Clarke starts answering about five times sooner and finishes sooner, while Sonnet scored higher on accuracy and writes slightly faster once it starts. GPT-5.4 results on the same tasks are pending.
GPT-5.4 pending
GPT-5.4 pending
GPT-5.4 pending
GPT-5.4 pending
The infrastructure, measured by others.
One model ID, and the whole stack under it.
- PlatformThe Radium API, its router and RL engine, and a pool of tuned open-weight models.OpenAI- and Anthropic-compatible, so your SDK and prompts stay where they are.The router and RL engine choose the model for each request.Radium's model curation team selects and tunes the models.
- InfraDistributed computing across data center sites, on modified server firmware and Radium's own switch software.
- HardwareRacks of GPU servers in data centers across several sites.
Switching takes three values.
Point the SDK you already use at Radium with a new base URL, a Radium key and the model name. Switching back is the same change in reverse.
- Endpoints
- OpenAI- and Anthropic-compatible
- Context window
- 200,000 tokens
- Input
- Text and images
- Structured output
- JSON mode and JSON schema
- Tool calling
- Parallel calls, strict schemas, multi-step loops
- Streaming
- OpenAI and Anthropic formats, with tool call deltas
- Rate limits
- 1,000 requests a minute, burst of 33, no tokens-per-minute limit
- Coming later
- Responses API, embeddings and batch
Start on Clarke and send the hard requests to Hal.
Hal
Reasoning across sources, code and long agent loops.
Clarke
RAG, copilots, tool calling and most production traffic.
Tycho
Classification, extraction and routing at volume, on text.
Questions about Clarke.
How is Clarke built?
Radium builds Clarke from open-weight models that its model curation team selects and tunes, and serves them behind one API with a router and a reinforcement learning engine that choose the model for each request.
How do I move a workload to Clarke?
Set the base URL to https://api.radium.cloud/v1, use a Radium API key and set the model to clarke-1.0. Your SDK, prompts and tool schemas stay the same.
What does Clarke cost?
$0.75 per million input tokens and $4.50 per million output tokens, and cached input is $0.075. New accounts get $25 of credit with no card needed.
Does Radium train on my data?
Your prompts and outputs are never used to train a model.
How was Clarke benchmarked?
On Radium's own suite of 80 structured tasks with fixed answers and 40 tasks for speed, against Claude Sonnet, with GPT-5.4 results on the same suite pending. The full method, confidence intervals and limits are on the benchmark results page.
When should I use Hal instead?
Move your everyday traffic to Clarke.
One line of code to switch.
A different class of performance.
Every new account gets $25 in credit. No card required.
