Contact Contacts
AI ranking

Cheapest AI models by API price

The price of API access per million tokens, in dollars, by the cheapest offer across providers. Input and output are folded into one number: output is usually several times more expensive, and ranking models by the input price alone would put something other than the cheapest on top.

100 models in the table
1508 considered in the calculation
2 August 2026 measurement date at the source
27 July 2026 newest model released here
7 August 2026 last recalculated
# Model Developer Price, $ per 1M tokens Sign in$ / 1M Output$ / 1M Contexttokens
1 Llama 3.2 1b Instruct Meta AI 0.01 $0.01 $0.01 128K
2 Ling 2.6 Flash Ant Group (Ling) 0.015 $0.01 $0.03 262K
3 Google Gemma 2 Google DeepMind 0.015 $0.01 $0.03 8K
4 Qwen2.5-Coder 7B Instruct Alibaba (Qwen) 0.015 $0.01 $0.03 131K
5 Qwen2.5 Coder 3B Instruct Alibaba (Qwen) 0.015 $0.01 $0.03 33K
6 Qwen2.5 Coder 7B Alibaba (Qwen) 0.015 $0.01 $0.03 33K
7 Llama 3.2 3B Instruct Meta AI 0.02 $0.02 $0.02 131K
8 Llama 3.1 8B Meta AI 0.021 $0.02 $0.03 131K
9 Mistral Nemo Mistral AI 0.022 $0.15 $0.15 131K
10 Mistral Nemo Mistral AI 0.023 $0.02 $0.03 131K
11 Meta Llama 3.1 8B Instruct Turbo Meta AI 0.023 $0.02 $0.03 128K
12 Gemma 3n E4b It Google DeepMind 0.025 $0.02 $0.04 33K
13 DeepSeek R1 Distill Llama 8B DeepSeek 0.025 $0.03 $0.03 33K
14 GPT OSS 20B OpenAI 0.028 $0.01 $0.07 131K
15 Hermes3 8B Nous Research 0.029 $0.03 $0.04 131K
16 LFM 7B Liquid AI 0.029 $0.03 $0.04 131K
17 Qwen3 4B Alibaba (Qwen) 0.03 $0.03 $0.03 128K
18 Qwen2 VL 7B Instruct Alibaba (Qwen) 0.03 $0.02 $0.06 131K
19 Meta: Llama 3 8B Instruct Meta AI 0.033 $0.03 $0.04 8K
20 Llama 3.1 8B Instruct Meta AI 0.035 $0.03 $0.05 128K
21 Ministral 3B Mistral AI 0.04 $0.04 $0.04 128K
22 Ministral 3B (latest) Mistral AI 0.04 $0.04 $0.04 128K
23 Sarvam 30B Sarvam AI 0.04 $0.02 $0.1 66K
24 Granite 4.0 H Micro IBM 0.04 $0.02 $0.11 131K
25 Qwen 3 8B Alibaba (Qwen) 0.041 $0.18 $0.7 131K
26 Qwen2.5 7B Instruct Alibaba (Qwen) 0.043 $0.18 $0.7 131K
27 Gemma3 4B Google DeepMind 0.043 $0.03 $0.08 128K
28 Sao10K: Llama 3 8B Lunaris Community 0.043 $0.04 $0.05 8K
29 L3 8B Lunaris V1 Turbo Community 0.043 $0.04 $0.05 8K
30 Nemotron 3 Nano Omni 30B TEE NVIDIA 0.043 $0.02 $0.1 131K
31 Mistral Nemo Instruct 2407 TEE Mistral AI 0.043 $0.02 $0.1 131K
32 Nex N2 Mini Nex AGI 0.044 $0.03 $0.1 262K
33 Qwen2.5 Coder 7B fast Alibaba (Qwen) 0.045 $0.03 $0.09 32K
34 GPT OSS 120B OpenAI 0.048 $0.03 $0.14 131K
35 Qwen3.5 4B Alibaba (Qwen) 0.048 $0.04 $0.07 262K
36 Nova Micro Amazon 0.048 $0.03 $0.1 128K
37 Llama 3.2 11b Vision Instruct Meta AI 0.049 $0.05 $0.05 131K
38 Gemma 3 27B Google DeepMind 0.05 $0.03 $0.11 131K
39 Gemma 3 4B IT Google DeepMind 0.05 $0.04 $0.08 131K
40 Llama 3.2 3B Meta AI 0.05 $0.04 $0.08 128K
41 Gemma 4 E2B Google DeepMind 0.05 $0.04 $0.08 128K
42 Sao10K Stheno 8b Community 0.05 $0.05 $0.05 8K
43 Sao10k L3 8B Lunaris Community 0.05 $0.05 $0.05 8K
44 Mistral Devstral Small 2505 Mistral AI 0.053 $0.1 $0.3 128K
45 LFM2 24B A2B Liquid AI 0.053 $0.03 $0.12 33K
46 Mistral Nemo 12B Instruct Mistral AI 0.054 $0.04 $0.1 16K
47 Qwen Turbo Alibaba (Qwen) 0.055 $0.05 $0.2 1,000K
48 Gemma 3 12B IT Google DeepMind 0.055 $0.05 $0.1 131K
49 DeepSeek R1 Distill Llama 70B DeepSeek 0.055 $0.03 $0.13 131K
50 Qwen3.7 Flash Alibaba (Qwen) 0.055 $0.03 $0.13 1,000K
51 Manta Mini 1.0 MegaNova AI 0.055 $0.02 $0.16 8K
52 Manta Flash 1.0 MegaNova AI 0.055 $0.02 $0.16 16K
53 Meta Llama 3.1 8B Instant Meta AI 0.058 $0.05 $0.08 131K
54 Mistral: Mistral Small 3 Mistral AI 0.058 $0.05 $0.08 33K
55 Gemma 7B IT Google DeepMind 0.058 $0.05 $0.08 8K
56 Llama 3 8B Meta AI 0.058 $0.05 $0.08 8K
57 MythoMax 13B Community 0.06 $0.06 $0.06 4K
58 AutoGLM Phone 9B Multilingual Zhipu AI / Z.ai 0.061 $0.04 $0.14 66K
59 Amazon Nova Micro 1.0 Amazon 0.061 $0.04 $0.14 128K
60 Granite 4.1 8B IBM 0.063 $0.05 $0.1 131K
61 Qwen3 32B Alibaba (Qwen) 0.063 $0.7 $2.8 131K
62 Qwen3.5 9B Alibaba (Qwen) 0.063 $0.04 $0.15 262K
63 Llama 4 Scout 17B 16E Instruct Meta AI 0.063 $0.05 $0.1 10,000K
64 Llama 4 Maverick 17b 128e Instruct Meta AI 0.063 $0.05 $0.1 1,049K
65 Mellum2 12B A2.5B JetBrains 0.063 $0.05 $0.1 131K
66 Gemma 4 26B Google DeepMind 0.064 $0.03 $0.17 131K
67 Command R7B Arabic Cohere 0.066 $0.04 $0.15 128K
68 Command R7B Cohere 0.066 $0.04 $0.15 128K
69 GLM 4.6V FlashX Zhipu AI / Z.ai 0.068 $0.02 $0.21 200K
70 DeepSeek R1 0528 Qwen3 8B DeepSeek 0.068 $0.06 $0.09 128K
71 GLM Z1 Air Zhipu AI / Z.ai 0.07 $0.07 $0.07 32K
72 NVIDIA Nemotron Nano 9B v2 NVIDIA 0.07 $0.04 $0.16 131K
73 DeepSeek R1 Distill Qwen 14B DeepSeek 0.07 $0.07 $0.07 33K
74 Baichuan M2 32B Baichuan AI 0.07 $0.07 $0.07 131K
75 Sarvam 105B Sarvam AI 0.07 $0.04 $0.16 131K
76 Qwen Flash Alibaba (Qwen) 0.071 $0.05 $0.4 1,000K
77 Trinity Mini Arcee AI 0.071 $0.05 $0.15 131K
78 Qwen3 235B A22B Instruct 2507 Alibaba (Qwen) 0.072 $0.1 $0.1 262K
79 Granite 3.3 8B Instruct IBM 0.073 $0.03 $0.25 8K
80 DeepSeek Coder 6.7B DeepSeek 0.075 $0.06 $0.12 16K
81 CodeLlama 7B Meta AI 0.075 $0.06 $0.12 16K
82 Laguna XS 2.1 Poolside 0.075 $0.06 $0.12 262K
83 DeepSeek V4 Flash DeepSeek 0.078 $0.14 $0.28 1,049K
84 Qwen: Qwen3 235B A22B Instruct 2507 Alibaba (Qwen) 0.078 $0.07 $0.1 262K
85 Tongyi Intent Detect V3 Alibaba (Qwen) 0.08 $0.06 $0.14 8K
86 Phi 4 Multimodal Microsoft 0.08 $0.07 $0.11 128K
87 Phi 4 Microsoft 0.08 $0.06 $0.14 128K
88 Qwen3 30B A3B Instruct 2507 Alibaba (Qwen) 0.084 $0.05 $0.19 262K
89 Mistral 7B Instruct Mistral AI 0.085 $0.07 $0.28 33K
90 Llama 3.3 70B Instruct Meta AI 0.088 $0.05 $0.23 131K
91 Nemotron 3 Nano 30B A3B NVIDIA 0.088 $0.05 $0.2 262K
92 Qwen Turbo 2024-11-01 Alibaba (Qwen) 0.088 $0.05 $0.2 1,000K
93 Qwen Turbo 2025-04-28 Alibaba (Qwen) 0.088 $0.05 $0.2 1,000K
94 Qwen Turbo Latest Alibaba (Qwen) 0.088 $0.05 $0.2 1,000K
95 Mistral 7B Instruct V0.2 Mistral AI 0.088 $0.05 $0.25 33K
96 Qwen3 30B A3B Alibaba (Qwen) 0.088 $0.09 $0.2 131K
97 Command A Translate Cohere 0.09 $2.5 $10 8K
98 Mistral Small 3.2 24B Instruct Mistral AI 0.09 $0.06 $0.18 256K
99 Mistral Small (latest) Mistral AI 0.09 $0.06 $0.18 256K
100 DeepSeek R1 Distill Qwen 7B DeepSeek 0.09 $0.07 $0.14 33K

The table scrolls sideways: not all columns fit.

The price is the lowest among the model's providers, base tier, without batch or discounted rates. Input and output are shown separately on purpose: for most models the output costs several times more than the input, and the final bill depends on which of the two your task has more of.