Agentic AI
Estimate an agent session from the total billed input and output tokens across all its model calls.
What matters most
Later steps may resend context and add tool-result tokens. Include those repeated tokens and any retries in the session totals; cache eligibility and separately billed tools need their own checks.
Base example
Illustrative workload: 6,000 total billed input tokens and 3,000 total billed output tokens across one agent session, repeated for 3,000 sessions per month.
Low-cost agentic AI models in the base example
These rows are server-rendered, so search engines can read them before any script runs.
Cost-ranked options include only chat routes with catalog-confirmed function calling and complete input/output rates. Performance and availability are not guaranteed.
| Model | Input | Output | Monthly cost |
|---|---|---|---|
| Qwen2.5-Coder-7B | $0.0100 / 1M tokens | $0.0300 / 1M tokens | $0.45 |
| llama3.2-11b-vision-instruct | $0.0150 / 1M tokens | $0.0250 / 1M tokens | $0.49 |
| llama3.2-3b-instruct | $0.0150 / 1M tokens | $0.0250 / 1M tokens | $0.49 |
| Llama-3.2-3B-Instruct | $0.0200 / 1M tokens | $0.0200 / 1M tokens | $0.54 |
| Meta-Llama-3.1-8B-Instruct-Turbo | $0.0200 / 1M tokens | $0.0300 / 1M tokens | $0.63 |
| Mistral-Nemo-Instruct-2407 | $0.0200 / 1M tokens | $0.0400 / 1M tokens | $0.72 |
| hermes3-8b | $0.0250 / 1M tokens | $0.0400 / 1M tokens | $0.81 |
| lfm-7b | $0.0250 / 1M tokens | $0.0400 / 1M tokens | $0.81 |
Open the compare tool with the lowest-cost agentic AI candidates preselected.
Enter billed token totals across all calls to the same model. Context growth and retries are not predicted; other models, separately billed tools, and cache discounts are excluded. Calculate observed cost per accepted task from actual spend
Recommended cheaper models
| Model | Cost per session | Monthly cost |
|---|---|---|
| Qwen2.5-Coder-7B | $0.000150 | $0.45 |
| llama3.2-11b-vision-instruct | $0.000165 | $0.49 |
| llama3.2-3b-instruct | $0.000165 | $0.49 |
| Llama-3.2-3B-Instruct | $0.000180 | $0.54 |
| Meta-Llama-3.1-8B-Instruct-Turbo | $0.000210 | $0.63 |
| Mistral-Nemo-Instruct-2407 | $0.000240 | $0.72 |
| hermes3-8b | $0.000270 | $0.81 |
| lfm-7b | $0.000270 | $0.81 |
Open calculator to estimate your own workload cost with different token counts and model choices.
Browse providers to compare models from OpenAI, Anthropic, Google, Mistral, and other providers.
Pricing data from catalog last generated Nov 20, 2026. Verify before production decisions. Data sources.