Novita AI review

Novita AI earns its spot on price and catalog size. The serverless API covers 200+ models across text, image, audio and video, billed by the token, and the pricing page lists cache read rates for most current LLMs. OpenAI GPT OSS 120B costs $0.05 per 1M input tokens and $0.25 per 1M output, the lowest rate for that model on this list, and small models such as Llama 3.1 8B Instruct start at $0.02 input. Batch inference runs at an introductory 50% discount on supported models. Beyond serverless, Novita sells dedicated endpoints with isolated resources, GPU instances and an agent sandbox billed per second. Its prices are not lowest everywhere: DeepSeek V4 Flash 0731 is $0.44 per 1M input tokens here against $0.06 at DeepInfra. The site states 99.5% uptime, below Baseten's figure, and it publishes less detail on regions and reserved capacity than the top three.

Who it is for

Budget-minded developers who want very low per-token rates on popular open models and a large catalog.

Price

Pay per token; OpenAI GPT OSS 120B at $0.05 per 1M input tokens and $0.25 output.

OpenAI GPT OSS 120B 128K $0.05 /Mt $0.25 /Mt
Quoted from the vendor pricing page on November 7, 2026.

Key features

• Serverless APIs for 200+ text, image, audio and video models
• Cache read pricing on many LLMs
• Batch inference at an introductory 50% discount on supported models
• Dedicated endpoints with isolated resources
• Agent sandbox billed per second

Strengths and weaknesses

• Strength: GPT OSS 120B at $0.05 per 1M input tokens, the lowest rate for that model on this list
• Strength: Very large catalog, including older and smaller models
• Strength: GPU instances and sandboxes on the same account
• Weakness: Some rates are higher than rivals, such as DeepSeek V4 Flash 0731 at $0.44 input against $0.06 at DeepInfra
• Weakness: Less published detail on SLAs and regions than Fireworks or Baseten
• Weakness: Dedicated endpoint pricing is not on the main table

Key facts

• GPT OSS 120B: $0.05 input, $0.25 output per 1M tokens
• Models: 200+
• Batch: Introductory 50% discount
• Stated uptime: 99.5%

Where it ranks

Novita AI is No. 6 in our Best AI Inference Platforms verdict.