Top 20 lowest-priced models from frontier AI labs

OpenAI: gpt-oss-20b

gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B...

$0.02 per million input tokens 131,072 context$/M in 0.018$/M out 0.09openai

OpenAI: gpt-oss-20b (batch)

gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license. It uses a Mixture-of-Experts (MoE) architecture with 3.6B...

$0.02 per million input tokens 131,072 context$/M in 0.024$/M out 0.112openai

OpenAI: GPT-5 Nano (batch)

GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments....

$0.02 per million input tokens 400,000 context$/M in 0.025$/M out 0.2openai

Meta: Llama 3.2 1B Instruct

Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and...

$0.03 per million input tokens 60,000 context$/M in 0.027$/M out 0.201meta-llama

DeepSeek: DeepSeek V4.1 Flash

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It...

$0.03 per million input tokens 1,048,576 context$/M in 0.028$/M out 0.011deepseek

Mistral: Mistral Nemo

A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multilingual, supporting English, French,...

$0.03 per million input tokens 131,072 context$/M in 0.029$/M out 0.03mistralai

OpenAI: gpt-oss-120b (batch)

gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose...

$0.03 per million input tokens 131,072 context$/M in 0.03$/M out 0.136openai

Qwen: Qwen3.7 Flash

Qwen3.7 Flash is a vision-language reasoning model from Alibaba. It is suited for multimodal agents, visual coding, search, and computer interaction, with...

$0.03 per million input tokens 1,000,000 context$/M in 0.03$/M out 0.13qwen

Anthropic: Claude Haiku 5.5 (batch)

Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude...

$0.05 per million input tokens 1,000,000 context$/M in 0.05$/M out 0.25anthropic

OpenAI: GPT-6 Luna Pro (batch)

GPT-6 Luna Pro is the same underlying model as [GPT-6 Luna](https://openrouter.ai/openai/gpt-6-luna), served with `reasoning.mode` set to `pro` for...

$0.05 per million input tokens 1,050,000 context$/M in 0.05$/M out 0.25openai

Aussi dans: #9 AI models with the longest context window

OpenAI: GPT-6 Luna (batch)

GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive...

$0.05 per million input tokens 1,050,000 context$/M in 0.05$/M out 0.25openai

Aussi dans: #11 AI models with the longest context window

DeepSeek: DeepSeek V4.1 Flash (batch)

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It...

$0.05 per million input tokens 1,048,576 context$/M in 0.05$/M out 0.15deepseek

OpenAI: GPT-5 Nano

GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments....

$0.05 per million input tokens 400,000 context$/M in 0.05$/M out 0.4openai

Google: Gemini 2.5 Flash Lite (batch)

Gemini 2.5 Flash-Lite is a lightweight reasoning model in the Gemini 2.5 family, optimized for ultra-low latency and cost efficiency. It offers improved...

$0.05 per million input tokens 1,048,576 context$/M in 0.05$/M out 0.2google

OpenAI: GPT-4.1 Nano (batch)

For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size...

$0.05 per million input tokens 1,047,576 context$/M in 0.05$/M out 0.2openai

Google: Gemma 3 4B

Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140...

$0.05 per million input tokens 131,072 context$/M in 0.05$/M out 0.1google

Google: Gemma 3 12B

Gemma 3 introduces multimodality, supporting vision-language input and text outputs. It handles context windows up to 128k tokens, understands over 140...

$0.05 per million input tokens 131,072 context$/M in 0.05$/M out 0.15google

Mistral: Mistral Small 3

Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it...

$0.05 per million input tokens 32,768 context$/M in 0.05$/M out 0.08mistralai

Meta: Llama 3.1 8B Instruct

Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors. This 8B instruct-tuned version is fast and efficient. It has...

$0.05 per million input tokens 131,072 context$/M in 0.05$/M out 0.08meta-llama

Meta: Llama 3.2 3B Instruct

Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation,...

$0.05 per million input tokens 131,072 context$/M in 0.051$/M out 0.335meta-llama