od DeepSeekSzybki
DeepSeek V4 Flash
1M context, thinking + non-thinking modes, tool calls
deepseek-v4-flashGenerate chat completions with this model.
/api/v1/chat/completionsGET /api/v1/models?type=chat{
"model": "deepseek-v4-flash",
"messages": [
{
"role": "user",
"content": "Hello!"
}
],
"stream": false,
"thinking": true,
"max_tokens": 4096
}curl -X POST https://kunya.ai/api/v1/chat/completions \
-H "Authorization: Bearer kunya_YOUR_API_KEY" \
-H "Content-Type: application/json"
-d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Hello!"}],"stream":false,"thinking":true,"max_tokens":4096}'Updated 2026-10-06 · #2 performance · #3 value
DeepSeek
Oct 17, 2026 V4.1 Flash snapshot (CED MoE) — native vision, coding & agentic workloads, 1M context, compressed KV cache
Dokumentacja APIKunya
Fast Tab completion router — routes to the configured flash model with curation telemetry.
Dokumentacja APIInception
Fastest reasoning diffusion LLM — parallel token generation (~140 tps), tunable reasoning, tool calling, and JSON schema output. Built for search agents, voice pipelines, and coding subagents.
Dokumentacja APIMiniMax
M2.5 at ~100 tps — same performance, faster and more agile
Dokumentacja API