Alla modellerchatDeepSeek V4 Flash

DeepSeek V4 Flash

av DeepSeekSnabb

Testa på Kunya
chat

DeepSeek V4 Flash

1M context, thinking + non-thinking modes, tool calls

Modell-ID:deepseek-v4-flash

API-dokumentation

Generate chat completions with this model.

POST/api/v1/chat/completions
GET /api/v1/models?type=chat
Request body
{
  "model": "deepseek-v4-flash",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "stream": false,
  "thinking": true,
  "max_tokens": 4096
}
cURL-exempel
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}'

Anteckningar

  • •OpenAI-compatible endpoint — works with any Kunya API key.
  • •Set `"stream": true` for server-sent events.
  • •Supports tool calling via the `tools` parameter.

Priser

Indata$0.286 per 1M tokens
Utdata$0.858 per 1M tokens
Kontextfönster1000K

Kunya benchmark

Updated 2026-10-06 · #2 performance · #3 value

Performance
97.5
Value (KVI)
120.7
Literary writing
95.5
Hard tier
77.1
SWE Verified
88.8%
Cost
$0.92/1k tasks
Full benchmark report →

Funktioner

Streaming Ja
Vision Nej
Resonemang Ja
Verktyg Ja
LeverantörDeepSeek
Testa på Kunya

Rankningar

Overall#4
Programming#7

Liknande modeller

DeepSeek V4.1 Flash 1017

DeepSeek

Oct 17, 2026 V4.1 Flash snapshot (CED MoE) — native vision, coding & agentic workloads, 1M context, compressed KV cache

API-dokumentation

DeepSeek V4 Pro

DeepSeek

Flagship model — 1M context, thinking + non-thinking modes

API-dokumentation

MiniMax M2.5 Highspeed

MiniMax

M2.5 at ~100 tps — same performance, faster and more agile

API-dokumentation

Nemotron 3.5 Lightning

NVIDIA

30B MoE (3B active) — high-throughput agentic workloads, tool calling, 1M context. Free on OpenRouter (rate-limited). Newer/fast tier vs Nemotron 3 Nano.

API-dokumentation