Powerful open-source model excelling at code and mathematics.
| Provider | Role | Context | Price · in / out | p50 | p95 | Uptime |
|---|---|---|---|---|---|---|
| ggmicloudNo trainingNo training on your data | Primary | 164K | $0.44 / $0.58/MToksame price, any rail | 2.2s | 3.5s | 100% |
| ddeepinfraNo trainingNo training on your data | Fallback | 164K | 6.6s | 6.6s | 100% |
Collecting — charts appear after two days.
curl https://www.ninjachat.ai/api/v1/chat/completions \
-H "Authorization: Bearer $NINJACHAT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v3.2",
"messages": [{ "role": "user", "content": "Hello!" }],
"stream": true
}'DeepSeek's widely used V3.2 model on two independently operated neutral hosts.
DeepSeek V3.2 costs $0.44/M input tokens and $0.58/M output tokens.
DeepSeek V3.2 accepts text and returns text.
POST /api/v1/chat/completions with `"model": "deepseek-v3.2"`.
DeepSeek V3, DeepSeek V4 Flash 0731, DeepSeek V4 Pro 0813.
routing.strategyproviders.onlyproviders.excludeproviders.ordermax_cost_usddata_policydocs →DeepSeek's production V4 Pro release on a no-training neutral inference host.
Anthropic's highest-capability generally available model for long-running agents.
Anthropic's fastest Claude variant — lightning-quick responses.
Anthropic's most intelligent model — use when quality is paramount.