LLMRates.ai
DeepInfra logo

DeepInfra API pricing

Model Aggregator

DeepInfra is a serverless inference marketplace running open models (Llama, Qwen, DeepSeek, GLM, Kimi and more) on managed GPUs, billed per token with no idle cost — among the cheapest hosted open-model APIs.

Compare the published API prices listed for DeepInfra. Open a model for its available pricing tiers, source and conditions, or use the cost planner to estimate your own workload.

Models
183
Prompt Caching
✓
Batch Discount
—
Tool Calling
✓
Visit Website →View Official Pricing →
Model183
anthropic/claude-fable-5$10$50
anthropic/claude-haiku-4-5$1$5
anthropic/claude-opus-4-7$5$25
anthropic/claude-opus-4-8$5$25
anthropic/claude-opus-5$5$25
anthropic/claude-opus-5-5$4$20
anthropic/claude-sonnet-4-6$3$15
anthropic/claude-sonnet-5$3$15
anthropic/claude-sonnet-5-5$2$10
Audio8/Audio8-TTS-Preview-0.6b$5/1M chars—
BAAI/bge-base-en-v1.5$0.005—
BAAI/bge-en-icl$0.01—
BAAI/bge-large-en-v1.5$0.01—
BAAI/bge-m3$0.01—
BAAI/bge-m3-multi$0.01—
BAAI/bge-m3-multi-8k$0.01—
bosonai/HiggsAudioV2.5$20/1M chars—
ByteDance/Seed-1.8$0.25$2
ByteDance/Seed-2.0-code$0.5$3
ByteDance/Seed-2.0-mini$0.1$0.4
ByteDance/Seed-2.0-pro$0.5$3
ByteDance/Seedance-1.5-Pro$1.2—
ByteDance/Seedance-2.0$4.7—
canopylabs/orpheus-3b-0.1-ft$7/1M chars—
Cloudflare/clef$0.24—
Cloudflare/clef-flash$0.038—
deepseek-ai/DeepSeek-R1-0528DeepSeek$0.5$2.15
deepseek-ai/DeepSeek-V3DeepSeek$0.32$0.89
deepseek-ai/DeepSeek-V3-0324DeepSeek$0.24$0.9
deepseek-ai/DeepSeek-V3.1DeepSeek$0.25$0.95
deepseek-ai/DeepSeek-V3.1-TerminusDeepSeek$0.27$0.95
deepseek-ai/DeepSeek-V3.2DeepSeek$0.26$0.38
deepseek-ai/DeepSeek-V4-FlashDeepSeek$0.09$0.18
deepseek-ai/DeepSeek-V4-Flash-0731DeepSeek$0.06$0.18
deepseek-ai/DeepSeek-V4-Flash-Vision-ExpDeepSeek$0.44$1.32
deepseek-ai/DeepSeek-V4-ProDeepSeek$1.3$2.6
deepseek-ai/DeepSeek-V4-Pro-0813DeepSeek$1.3$2.6
deepseek-ai/DeepSeek-V4.1-FlashDeepSeek$0.2$0.6
deepseek-ai/DeepSeek-V4.1-Flash-1017DeepSeek$0.2$0.6
google/embeddinggemma-2Gemma$0.02—
google/embeddinggemma-300mGemma$0.002—
google/gemini-2.5-flash$0.3$2.5
google/gemini-2.5-pro$1.25$10
google/gemini-3-pro-image$120—
google/gemini-3.1-flash-lite$0.25$1.5
google/gemini-3.1-pro$2$12
google/gemini-3.5-flash$1.5$9
google/gemini-3.7-flash$0.75$3.75
google/gemma-3-12b-itGemma$0.05$0.15
google/gemma-3-27b-itGemma$0.08$0.16
Showing 1–50 of 183

DeepInfra API cost examples

1,000 requests, each using 1,000 input tokens and 200 billed text output tokens: 1M input and 200K output tokens in total.

Standard text token prices and estimated workload cost in USD
ModelInput / 1M tokensOutput / 1M tokensEstimated costPricing conditions
meta-llama/Meta-Llama-3.1-8B-Instruct-TurboDeepInfra$0.02$0.04$0.028Region not specifiedView official pricing
mistralai/Mistral-Nemo-Instruct-2407DeepInfra$0.029$0.03$0.035Region not specifiedView official pricingVerified Oct 14, 2026
Sao10K/L3-8B-Lunaris-v1-TurboDeepInfra$0.04$0.05$0.05Region not specifiedView official pricing
openai/gpt-oss-20bDeepInfra$0.03$0.14$0.058Region not specifiedView official pricing
mistralai/Mistral-Small-24B-Instruct-2501DeepInfra$0.05$0.08$0.066Region not specifiedView official pricing
google/gemma-3-4b-itDeepInfra$0.05$0.1$0.07Region not specifiedView official pricing

Estimates in USD use standard processing, uncached input and the displayed base input tier. Regional rates, longer prompts, thinking tokens and other billable usage can change the final cost. Models can differ in capabilities.

Calculate costs for your workload →