DeepInfra pricing

62 text models on DeepInfra, from $0.02 to $2.85 per 1M input tokens.

Updated

Running an agent on these models? abracadabra shows which tasks it finishes and where the rest stop.

62 models

MiMo-V2.6-Pro

Xiaomi1.05M contextText and image inputAdded 21 Sept 2026

Input
$0.387
Output
$0.783
Cached
$0.0032
Batch
–

DeepSeek V4 Flash Vision Exp

DeepSeek1.05M contextText and image inputAdded 21 Aug 2026

Input
$0.2156
Output
$0.6468
Cached
$0.0069
Batch
–

GLM 5.3

Z.ai1.05M contextText inputAdded 18 Aug 2026

Input
$0.5625
Output
$2.50
Cached
$0.125
Batch
−20%

Muse Glimmer 30B

Meta131K contextText and image inputAdded 9 Aug 2026

Input
$0.30
Output
$1.20
Cached
$0.04
Batch
−20%

Inkling Small

Thinking Machines524K contextText and image inputAdded 30 Jul 2026

Input
$0.45
Output
$1.20
Cached
$0.10
Batch
−20%

Kimi K3

Moonshot AI1.05M contextText and image inputAdded 16 Jul 2026

Input
$2.85
Output
$14.25
Cached
$0.285
Batch
−20%

Nemotron 3.5 Content Safety

NVIDIA131K contextText and image inputAdded 4 Jun 2026

Input
$0.20
Output
$0.20
Cached
–
Batch
–

MiniMax M3

MiniMax524K contextText and image inputAdded 31 May 2026

Input
$0.224
Output
$0.88
Cached
$0.0448
Batch
–

Kimi K2.6

Moonshot AI262K contextText and image inputAdded 20 Apr 2026

Input
$0.75
Output
$3.50
Cached
$0.15
Batch
–

Nemotron 3 Super

NVIDIA262K contextText inputAdded 11 Mar 2026

Input
$0.075
Output
$0.40
Cached
–
Batch
–

Qwen3 VL 235B A22B Instruct

Qwen262K contextText and image inputAdded 23 Sept 2025

Input
$0.20
Output
$0.88
Cached
$0.11
Batch
–

Qwen3 Next 80B A3B Instruct

Qwen262K contextText inputAdded 11 Sept 2025

Input
$0.072
Output
$0.88
Cached
–
Batch
–

Qwen3 Coder 480B A35B

Qwen262K contextText inputAdded 23 Jul 2025

Input
$0.30
Output
$1.00
Cached
$0.10
Batch
–

R1 0528

DeepSeek164K contextText inputAdded 28 May 2025

Input
$0.50
Output
$2.15
Cached
$0.35
Batch
–

Llama Guard 4 12B

Meta164K contextText and image inputAdded 30 Apr 2025

Input
$0.144
Output
$0.144
Cached
–
Batch
–

Qwen3 30B A3B

Qwen41K contextText inputAdded 28 Apr 2025

Input
$0.12
Output
$0.50
Cached
–
Batch
–

Qwen3 14B

Qwen41K contextText inputAdded 28 Apr 2025

Input
$0.12
Output
$0.24
Cached
–
Batch
–

Qwen3 32B

Qwen41K contextText inputAdded 28 Apr 2025

Input
$0.08
Output
$0.28
Cached
–
Batch
–

Gemma 3 4B

Google131K contextText and image inputAdded 13 Mar 2025

Input
$0.05
Output
$0.10
Cached
–
Batch
–

Gemma 3 12B

Google131K contextText and image inputAdded 13 Mar 2025

Input
$0.04
Output
$0.12
Cached
–
Batch
–

Mistral Small 3

Mistral33K contextText inputAdded 30 Jan 2025

Input
$0.05
Output
$0.08
Cached
–
Batch
–

Phi 4

Microsoft16K contextText inputAdded 10 Jan 2025

Input
$0.056
Output
$0.112
Cached
–
Batch
–

DeepSeek V3

DeepSeek164K contextText inputAdded 26 Dec 2024

Input
$0.256
Output
$0.712
Cached
–
Batch
–

Qwen2.5 72B Instruct

Qwen33K contextText inputAdded 19 Sept 2024

Input
$0.36
Output
$0.40
Cached
–
Batch
–

Prices in US dollars. Token prices are per 1 million tokens. abracadabra compiles them from each provider’s listing on OpenRouter, the LiteLLM price list and the providers’ own pricing pages, and refreshes them daily. The dataset is free to use: /api/ai-pricing.

All DeepInfra models

Text models

See which tasks your agent finishes.

A price says what a call costs. abracadabra reads every conversation your agent has and shows the topics people bring, the tasks that get done and the turn where the rest stop.

Free to start, no card. Or open the playground.