Baseten review

Baseten covers two jobs that many teams need at once. Model APIs serve popular open models per token, with GPT OSS 120B at $0.10 per 1M input tokens and $0.50 per 1M output tokens, the lowest rate for that model in this list, plus Fast variants of GLM and DeepSeek models for latency-sensitive work at a higher price. Dedicated Inference runs your own models on GPUs billed per minute, from a T4 at about $0.63 an hour to an H100 at $6.50 and a B200 at $9.98, and training uses the same rates. Baseten says each model is checked against the lab's published benchmarks and runs at the precision it publishes. Pay as you go needs no contract, while volume discounts, reserved compute and custom regions go through sales. Dedicated deployments need more setup than a serverless call, which is why it scores below the top three on ease of use.

Who it is for

Teams serving both popular open models and their own models

Price

Pay as you go; GPT OSS 120B at $0.10 per 1M input tokens and $0.50 per 1M output tokens; dedicated H100 at $6.50 per hour billed per minute.

Model GPT OSS 120B Input $0.10 Cache input — Output $0.50
Quoted from the vendor pricing page on November 12, 2026.

Key features

• Model APIs that serve new open models from launch day, billed per input, cached input and output token
• Fast variants of GLM and DeepSeek models at a higher token price
• Dedicated Inference on T4 to B200 GPUs billed per minute
• Training on the same per-minute GPU rates
• Sandboxes for agents billed per vCPU hour and GB of RAM

Strengths and weaknesses

• Strength: Lowest listed gpt-oss-120B price in this list at $0.10 input and $0.50 output
• Strength: H100 MIG with 40 GiB at $3.75 per hour for smaller models
• Strength: States that models run at the published precision with no silent downgrades
• Weakness: Fast variants cost more, for example GLM-5.3 Fast at $2.10 input versus $1.40
• Weakness: Volume discounts and reserved compute require talking to sales
• Weakness: Dedicated deployments take more setup than calling a serverless endpoint

Key facts

• GPT OSS 120B: $0.10 input, $0.50 output per 1M tokens
• Dedicated H100: $6.50 per hour, billed per minute
• Dedicated B200: $9.98 per hour
• Billing: Usage-based, no contracts on pay as you go

Where it ranks

Baseten is No. 4 in our Best AI Inference Platforms verdict.