DeepInfra models and pricing

We track 57 models on DeepInfra (58 listings) from 14 labs. The cheapest paid model is Mistral Small 3, at $0.05 per million input tokens and $0.08 per million output tokens. Input prices range from $0.03 to $10 and output prices from $0.08 to $50 per million tokens. The largest context window is 1M tokens (Inkling 2 Small, Kimi K3.1 Pro Highspeed and 13 other models). Prices last verified 18 Nov 2026.

Prices in USD per million tokens. Official site

Cheapest models on DeepInfra

Input / output per million tokens, paid models, by combined price.

Newest models on DeepInfra

By the lab's release date.

All DeepInfra models and prices

ModelLabInputOutputContextSource
Qwen 4 Max
Qwen/Qwen4-Max
Alibaba (Qwen)$1.41$4.241Mmodels.dev
Inkling 2 Small
thinkingmachines/Inkling-2-Small
Thinkingmachines$0.30$11Mmodels.dev
Kimi K3.1 Pro Highspeed
moonshotai/Kimi-K3.1-Pro
Moonshot AI$2$101Mmodels.dev
DeepSeek V4.1 Flash 1017
deepseek-ai/DeepSeek-V4.1-Flash-1017
DeepSeek$0.20$0.601Mmodels.dev
MiMo-V2.6-Flash
XiaomiMiMo/MiMo-V2.6-Flash
Xiaomi (MiMo)$0.14$0.281Mmodels.dev
MiMo-V2.6-Pro
XiaomiMiMo/MiMo-V2.6-Pro
Xiaomi (MiMo)$0.43$0.871Mmodels.dev
DeepSeek V4.1 Flash
deepseek-ai/DeepSeek-V4.1-Flash
DeepSeek$0.20$0.601Mmodels.dev
Tencent Hy4 Preview
tencent/Hy4-preview
Tencent (Hunyuan)$0.83$2.501Mmodels.dev
GLM-5.3-Flash
zai-org/GLM-5.3-Flash
Zhipu AI (Z.ai)$0.15$0.501Mmodels.dev
Qwen3.8 Flash
Qwen/Qwen3.8-Flash
Alibaba (Qwen)$0.11$0.381Mmodels.dev
DeepSeek V4 Flash Vision Exp
deepseek-ai/DeepSeek-V4-Flash-Vision-Exp
DeepSeek$0.44$1.321Mmodels.dev
GLM-5.3
zai-org/GLM-5.3
Zhipu AI (Z.ai)$0.90$41Mmodels.dev
Qwen3.8 27B
Qwen/Qwen3.8-27B
Alibaba (Qwen)$0.20$2.50262Kmodels.dev
DeepSeek V4 Pro
deepseek-ai/DeepSeek-V4-Pro-0813
DeepSeek$1.30$2.601Mmodels.dev
DeepSeek V4 Pro
deepseek-ai/DeepSeek-V4-Pro
DeepSeek$1.30$2.601Mmodels.dev
Qwen3.8 Max
Qwen/Qwen3.8-Max
Alibaba (Qwen)$1.65$4.95256Kmodels.dev
DeepSeek V4 Flash 0731
deepseek-ai/DeepSeek-V4-Flash-0731
DeepSeek$0.06$0.181Mmodels.dev
Inkling Small
thinkingmachines/Inkling-Small
Thinkingmachines$0.45$1.20524Kmodels.dev
Kimi K3
moonshotai/Kimi-K3
Moonshot AI$2.85$14.251Mmodels.dev
GLM-5.2
zai-org/GLM-5.2
Zhipu AI (Z.ai)$0.75$2.401Mmodels.dev
Claude Fable 5
anthropic/claude-fable-5
Anthropic$10$501Mmodels.dev
MiniMax-M3
MiniMaxAI/MiniMax-M3
MiniMax$0.28$1.10524Kmodels.dev
Qwen3.7 Max
Qwen/Qwen3.7-Max
Alibaba (Qwen)$2.50$7.50256Kmodels.dev
DeepSeek V4 Flash
deepseek-ai/DeepSeek-V4-Flash
DeepSeek$0.09$0.181Mmodels.dev
Qwen3.6 27B
Qwen/Qwen3.6-27B
Alibaba (Qwen)$0.32$3.20262Kmodels.dev
Kimi K2.6
moonshotai/Kimi-K2.6
Moonshot AI$0.75$3.50262Kmodels.dev
Qwen3.6 35B-A3B
Qwen/Qwen3.6-35B-A3B
Alibaba (Qwen)$0.10$0.95262Kmodels.dev
Gemma 4 26B A4B IT
google/gemma-4-26B-A4B-it
Google DeepMind$0.07$0.34262Kmodels.dev
Gemma 4 31B IT
google/gemma-4-31B-it
Google DeepMind$0.20$0.40262Kmodels.dev
Qwen/Qwen3.5-9B
Qwen/Qwen3.5-9B
Alibaba (Qwen)$0.10$0.15262Kmodels.dev
Qwen3.5 27B
Qwen/Qwen3.5-27B
Alibaba (Qwen)$0.26$2.60262Kmodels.dev
Qwen3.5 35B-A3B
Qwen/Qwen3.5-35B-A3B
Alibaba (Qwen)$0.14$1262Kmodels.dev
Qwen3.5 397B-A17B
Qwen/Qwen3.5-397B-A17B
Alibaba (Qwen)$0.45$3262Kmodels.dev
Seed 2.0 Mini
ByteDance/Seed-2.0-mini
ByteDance Seed$0.10$0.40256Kmodels.dev
Seed 2.0 Code
ByteDance/Seed-2.0-code
ByteDance Seed$0.50$3256Kmodels.dev
Seed 2.0 Pro
ByteDance/Seed-2.0-pro
ByteDance Seed$0.50$3256Kmodels.dev
GLM-4.7
zai-org/GLM-4.7
Zhipu AI (Z.ai)$0.40$1.75203Kmodels.dev
DeepSeek V3.2
deepseek-ai/DeepSeek-V3.2
DeepSeek$0.26$0.38164Kmodels.dev
Qwen3 VL 30B A3B Instruct
Qwen/Qwen3-VL-30B-A3B-Instruct
Alibaba (Qwen)$0.15$0.60262Kmodels.dev
GLM-4.6
zai-org/GLM-4.6
Zhipu AI (Z.ai)$0.50$2203Kmodels.dev
Qwen3 VL 235B A22B Instruct
Qwen/Qwen3-VL-235B-A22B-Instruct
Alibaba (Qwen)$0.20$0.88262Kmodels.dev
Qwen3-Next 80B-A3B Instruct
Qwen/Qwen3-Next-80B-A3B-Instruct
Alibaba (Qwen)$0.09$1.10262Kmodels.dev
DeepSeek-V3.1
deepseek-ai/DeepSeek-V3.1
DeepSeek$0.25$0.95164Kmodels.dev
gpt-oss-20b
openai/gpt-oss-20b
OpenAI$0.03$0.14131Kmodels.dev
gpt-oss-120b
openai/gpt-oss-120b
OpenAI$0.04$0.17131Kmodels.dev
Qwen3 235B A22B Instruct 2507
Qwen/Qwen3-235B-A22B-Instruct-2507
Alibaba (Qwen)$0.09$0.55262Kmodels.dev
DeepSeek-R1-0528
deepseek-ai/DeepSeek-R1-0528
DeepSeek$0.50$2.15164Kmodels.dev
Llama Guard 4 12B
meta-llama/Llama-Guard-4-12B
Meta$0.18$0.18164Kmodels.dev
Qwen3-30B-A3B
Qwen/Qwen3-30B-A3B
Alibaba (Qwen)$0.12$0.5041Kmodels.dev
Llama-4-Scout-17B-16E-Instruct-FP8
meta-llama/Llama-4-Scout-17B-16E-Instruct
Meta$0.10$0.30328Kmodels.dev
Qwen3-Coder 480B-A35B Instruct
Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo
Alibaba (Qwen)$0.30$1262Kmodels.dev
Qwen3 32B
Qwen/Qwen3-32B
Alibaba (Qwen)$0.08$0.2841Kmodels.dev
Gemma 3 27B IT
google/gemma-3-27b-it
Google DeepMind$0.08$0.16131Kmodels.dev
Gemma 3 4B IT
google/gemma-3-4b-it
Google DeepMind$0.05$0.10131Kmodels.dev
Gemma 3 12B IT
google/gemma-3-12b-it
Google DeepMind$0.05$0.15131Kmodels.dev
Mistral Small 3
mistralai/Mistral-Small-24B-Instruct-2501
Mistral AI$0.05$0.0833Kmodels.dev
DeepSeek-V3
deepseek-ai/DeepSeek-V3
DeepSeek$0.32$0.89164Kmodels.dev
Llama-3.3-70B-Instruct
meta-llama/Llama-3.3-70B-Instruct-Turbo
Meta$0.10$0.32131Kmodels.dev

Frequently asked questions

How much does DeepInfra cost per million tokens?

Prices on DeepInfra depend on the model. Across the 57 models we track, input costs $0.03 to $10 per million tokens and output costs $0.08 to $50 per million tokens. Prices last verified 18 Nov 2026.

What is the cheapest model on DeepInfra?

Mistral Small 3 from Mistral AI is the cheapest paid model we track on DeepInfra, at $0.05 per million input tokens and $0.08 per million output tokens. Next are Gemma 3 4B IT ($0.05 / $0.10) and gpt-oss-20b ($0.03 / $0.14). Prices last verified 18 Nov 2026.

How many models does DeepInfra offer?

We currently track 57 active models on DeepInfra from 14 labs, led by Alibaba (Qwen) (18), DeepSeek (10), Google DeepMind (5) and Zhipu AI (Z.ai) (5).

Which model on DeepInfra has the largest context window?

Inkling 2 Small, Kimi K3.1 Pro Highspeed and 13 other models share the largest context window we track on DeepInfra: 1,048,576 tokens (1M).