Skip to content
NewTry RouterPlus with free creditsTry playground →
RouterPlus
Book a demo

DeepInfra API

Open in consoleI need inference
  • 27Models
  • 14Labs
  • 1.05MMax context
  • $0.03/ $0.14From, per 1M in / out

Pricing

ModelLabContextInput / 1MCached / 1MOutput / 1MProviders
Nemotron 3.5 Super VLnvidia/nemotron-3.5-super-vlNVIDIA262K$0.15$0.02$0.501
Ling 3.1 Flashinclusionai/ling-3.1-flashInclusionAI262K$0.075$0.015$0.2252
MiMo V2.6 Flashxiaomi/mimo-v2.6-flashXiaomi1.05M$0.14$0.0028$0.282
MiMo V2.6 Proxiaomi/mimo-v2.6-proXiaomi1.05M$0.387$0.0032$0.7832
DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flashDeepSeek1.05M$0.14$0.0042$0.429
Ling 3.0 Flash VLinclusionai/ling-3.0-flash-vlInclusionAI131K$0.06$0.012$0.181
Qwen3.8 Maxqwen/qwen3.8-maxQwen256K$1.65$0.206$4.9511
Granite 4.2 8Bibm-granite/granite-4.2-8bIBM131K$0.06$0.015$0.251
Hy4 Previewtencent/hy4-previewTencent1.05M$0.834$0.042$2.5012
GLM 5.3 Flashz-ai/glm-5.3-flashZ.ai1.05M$0.15$0.03$0.5012
Qwen3.8 Flashqwen/qwen3.8-flashQwen1M$0.113$0.0141$0.3821
DeepSeek V4 Flash Vision Expdeepseek/deepseek-v4-flash-vision-expDeepSeek1.05M$0.2156$0.0069$0.64681
GLM 5.3z-ai/glm-5.3Z.ai1.05M$0.5625$0.125$2.5012
DeepSeek V4 Pro 0813deepseek/deepseek-v4-pro-0813DeepSeek1.05M$1.30$0.10$2.606
Muse Glimmer 30Bmeta/muse-glimmer-30bMeta131K$0.30$0.04$1.202
DeepSeek V4 Flash 0731deepseek/deepseek-v4-flash-0731DeepSeek1.05M$0.06$0.015$0.189
Inkling Smallthinkingmachines/inkling-smallThinking Machines524K$0.45$0.10$1.201
Ling 3.0 Flashinclusionai/ling-3.0-flashInclusionAI131K$0.06$0.012$0.182
Kimi K3moonshotai/kimi-k3Moonshot AI1.05M$2.85$0.285$14.2510
GLM 5.2z-ai/glm-5.2Z.ai1.05M$0.5625$0.105$1.808
MiniMax M3minimax/minimax-m3MiniMax524K$0.28$0.056$1.103
Qwen3.6 27Bqwen/qwen3.6-27bQwen262K$0.32–$3.204
Qwen3.6 35B A3Bqwen/qwen3.6-35b-a3bQwen262K$0.10$0.10$0.954
DeepSeek V4 Prodeepseek/deepseek-v4-proDeepSeek1.05M$1.30$0.10$2.604
Gemma 4 31Bgoogle/gemma-4-31b-itGoogle262K$0.20–$0.404
gpt-oss-120bopenai/gpt-oss-120bOpenAI131K$0.037–$0.172
gpt-oss-20bopenai/gpt-oss-20bOpenAI131K$0.03–$0.141

DeepInfra's price, USD, as of November 5, 2026. Live prices

Quickstart

OpenAI-compatible
curl https://api.routerplus.com/v1/chat/completions \
  -H "Authorization: Bearer $ROUTERPLUS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "z-ai/glm-5.3",
    "messages": [{ "role": "user", "content": "Hello" }],
    "max_tokens": 1024,
    "provider": { "only": ["deepinfra-chat"] }
  }'

FAQ

How do I send a request to DeepInfra?

Routing tries providers cheapest first, by input plus output price, so DeepInfra serves a model where it is the cheapest and otherwise as a fallback. To send a request to DeepInfra first, name it in provider.only or provider.order as "deepinfra-chat".

How much does DeepInfra cost on RouterPlus?

You pay DeepInfra's own price for the model. DeepInfra's lowest input price is $0.03 per million tokens (gpt-oss-20b) and its lowest output price is $0.14 per million tokens (gpt-oss-20b). Prices as of November 5, 2026.

Which models does DeepInfra serve?

Nemotron 3.5 Super VL, Ling 3.1 Flash, MiMo V2.6 Flash, MiMo V2.6 Pro, DeepSeek V4.1 Flash, Ling 3.0 Flash VL, Qwen3.8 Max, Granite 4.2 8B, Hy4 Preview, GLM 5.3 Flash, Qwen3.8 Flash, DeepSeek V4 Flash Vision Exp, GLM 5.3, DeepSeek V4 Pro 0813, Muse Glimmer 30B, DeepSeek V4 Flash 0731, Inkling Small, Ling 3.0 Flash, Kimi K3, GLM 5.2, MiniMax M3, Qwen3.6 27B, Qwen3.6 35B A3B, DeepSeek V4 Pro, Gemma 4 31B, gpt-oss-120b and gpt-oss-20b.

Does DeepInfra keep my prompts?

DeepInfra declares no prompt retention in its RouterPlus listing, per its published policy.

Does DeepInfra train on my prompts?

No. DeepInfra's data-privacy page says it does not train on API data, does not store inference data on disk and does not log request content. The gateway calls DeepInfra's OpenAI-compatible API directly, never its bulk inference API.

Are Claude and Gemini on DeepInfra too?

Yes, as a separate listing, deepinfra-labs, which DeepInfra relays to Anthropic and Google under their own data policies. It is Claude's fallback after Anthropic direct. This page covers DeepInfra's open models.

Route to DeepInfra with one key

Open in consoleI need inference