LongCat 2.0
sdn-longcat-2CodingLive in the catalogA reasoning-capable mixture-of-experts flagship built around an enormous context window. Its distinguishing feature is how cheaply it re-reads: repeated context counts at roughly two percent of fresh input, so the long stable prefix an agent resends every turn costs almost nothing to keep. That makes it a natural fit for work that carries a lot of context forward — long sessions, large document sets, sprawling codebases.
- 011M-token context window
- 02Deep cached-input discount — repeated context is nearly free
- 03Reasoning-capable, with tool use on long-running work
Strict, schema-faithful tool calls, enforced by the engine on every request — safe to build an agent loop on.
Thinks before it answers. Budget max_tokens generously — hidden reasoning counts against it.
Text-only.
One endpoint. Served by our engine.
LongCat 2.0 is served through the Sideren engine — zero-downtime serving is the design target, not a status-page apology. You request it by name; everything else is our problem.
curl https://api.sideren.io/v1/messages \
-H "x-api-key: sdn_your_key" \
-H "content-type: application/json" \
-d '{
"model": "sdn-longcat-2",
"max_tokens": 1024,
"messages": [{ "role": "user", "content": "Hello" }]
}'OpenAI-style clients work too — send the same model name to /v1/chat/completions with a Bearer key. See the docs for both dialects.
Your agent doesn't change.
Everything underneath does.
Start free · no card · Starter from $5/mo