LLM price comparison
& token cost calculator
Compare Omev AI, OpenAI, Claude and Gemini at your monthly volume. Enter tokens, article counts or a bill to see the cost on each side.
For batch, cached usage or a negotiated rate. Changes the selected provider’s rates; Omev uses its published prices.
Your monthly estimate
Updates as you typeOmevAI Pro
$1.875
per month · published rates
GPT-5.6 Terra
$5.00
per month · published rates
$3.125 lower token cost on OmevAI Pro (62.5%).
See the calculation
Monthly volume: 1,000,000 input + 250,000 output tokens.
OmevAI Pro
$0.625 input + $1.25 output = $1.875.
GPT-5.6 Terra
$2.00 input + $3.00 output = $5.00.
Each subtotal = tokens ÷ 1,000,000 × the relevant rate. Rates used: Omev $0.625 input / $5 output; GPT-5.6 Terra $2 input / $12 output. Totals use full precision; displayed figures are rounded.
Standard text token pricing only, unless you enter effective rates. Excludes long-context premiums, cache storage and writes, tool fees, taxes and review time. Check OpenAI pricing ↗
Test Omev AI →$5 free credit · 10 days · no card
Calculated in your browser. Shared links include your inputs.
Omev AI publishes this tool. The catalog is a selected set of text models, not the entire market. Omev appears first; all rates are linked to their source. Quality and task suitability need a separate evaluation.
LLM API prices per million tokens
Standard paid text rates in USD, checked September 15, 2026. Input and output are priced separately. The calculator above uses these rates unless you enter effective prices.
| Model and source | Input / 1M | Output / 1M |
|---|---|---|
| OmevAI LiteOmev AI | $0.15 | $1.25 |
| OmevAI ProOmev AI | $0.625 | $5.00 |
| GPT-5.6 Luna ↗OpenAI | $0.20 | $1.20 |
| GPT-5.6 Terra ↗OpenAI | $2.00 | $12.00 |
| GPT-5.6 Sol ↗OpenAI | $4.00 | $20.00 |
| GPT-6 Astra ↗OpenAI | $10.00 | $50.00 |
| GPT-5.4 · legacy ↗OpenAI | $2.50 | $15.00 |
| GPT-5.4 mini · legacy ↗OpenAI | $0.75 | $4.50 |
| GPT-5.4 nano · legacy ↗OpenAI | $0.20 | $1.25 |
| GPT-5 nano · legacy ↗OpenAI | $0.05 | $0.40 |
| Claude Sonnet 5 ↗Anthropic | $2.00 | $10.00 |
| Claude Haiku 4.5 ↗Anthropic | $1.00 | $5.00 |
| Claude Sonnet 4.6 · legacy ↗Anthropic | $3.00 | $15.00 |
| Gemini 3.8 Flash ↗Google · promotional rate* | $0.75 | $3.75 |
| Gemini 3.5 Flash-Lite ↗Google | $0.30 | $2.50 |
| Gemini 3.6 Flash · legacy ↗Google | $0.75 | $3.75 |
| Gemini 3.5 Flash · legacy ↗Google | $1.50 | $9.00 |
| Gemini 3.1 Flash-Lite · legacy ↗Google | $0.25 | $1.50 |
| Gemini 3.1 Pro Preview · legacy ↗Google | $2.00 | $12.00 |
* The Gemini 3.8 Flash rate shown applies through December 31, 2026. Google lists $1.50 input / $7.50 output per 1M tokens from January 1, 2027. See Google’s pricing schedule.
Excluded: long-context premiums, cache storage and writes, batch or cached-input discounts, regional charges, tools, media and taxes. An API subscription or app plan is not a token rate. Verify the exact model and service tier before budgeting.
How to calculate LLM token costs
Add the input and output costs for the period you want to price. The same token volume is used on both sides, so the result isolates the difference between rates.
(input tokens ÷ 1,000,000 × input price)
+ (output tokens ÷ 1,000,000 × output price)
Token counts
Use actual billed input and output usage. Include retries and every generation call if you want to model the whole workflow.
Articles
Multiply article count by input and output tokens per article. Replace the example assumptions with your own measured totals.
Monthly bill
Infer volume from spend and the share of tokens that are output. This needs the rate that actually produced the bill.
For cached and uncached input combined, an effective input rate can be calculated as input-token charges ÷ billed input tokens × 1,000,000. Keep storage and other non-token charges separate. Different tokenizers, response lengths and retry rates can change the cost of completing the same job.
Worked example: 500 articles a month
Assume 3,000 input and 2,000 output tokens per article: 1.5M input and 1M output tokens monthly. This is an illustrative generation budget, not a benchmark or the total cost of publishing.
Omev Pro
$5.9375
per month
GPT-5.6 Terra
$15.00
per month
Token cost difference
$9.0625
per month
Review, storage and additional calls are extra. See our LLM cost optimization guide for budgeting approved output.
Use this example in the calculator ↑LLM pricing calculator FAQ
How do I calculate LLM API costs?
Divide input tokens by one million and multiply by the input rate. Do the same for output tokens at the output rate, then add the two costs. Use billed usage counts, including billed reasoning tokens where applicable, rather than estimating tokens from the visible response.
Can I compare batch, cached-input or negotiated prices?
Yes. Enable ‘Use my effective input and output rates’ and enter the prices that apply to your workload. For mixed cached and uncached usage, divide the relevant input-token charges by total input tokens and multiply by one million. Account for cache writes and storage separately. Omev remains priced at its published rates.
Which LLM API is cheapest?
There is no universal winner for every workload. For example, GPT-5 nano has lower published input and output rates than Omev Lite in this selection. Other comparisons depend on the token mix, discounts and the Omev model you choose. Price alone does not establish output quality or the cost of a result your team approves.
Can I estimate cost from my monthly API bill?
Use Monthly bill mode and enter the share of tokens that are output. This infers volume from the selected rates. Remove non-token charges first, and enter effective rates if your bill includes discounts. Use Token counts when you know the actual usage or when a zero rate makes volume impossible to infer.
Will the same prompt cost this much on every model?
The calculator holds token volume constant to compare rates. Models can tokenize the same prompt differently, produce different response lengths and require different retries. Test a representative workload and include review time before treating a price difference as production savings.
Is the calculator free, and can I share a result?
The calculator is free and requires no signup or API key. Copy calculation link preserves the active mode, model, Omev tier and entered values in the URL fragment. Anyone with that link can read those values and restore the estimate.
Continue your comparison
Turn the estimate into a workload test
Run representative prompts on Omev AI. Compare finished output, actual token use and review effort before moving production volume.
Test Omev AI with $5 free credit →10 days · No card required
By Di Reshtei, Co-founder of OmevAI · X