Cost formula
Scenario cost = (uncached input / 1M × input $/1M) + (cached input / 1M × cache $/1M) + (output / 1M × output $/1M), × requests; optional batch multiplier when supported
Calculator
Price table & scenario cost desk
Scan catalog rates like a price desk, tick up to eight text models, enter input/cached/output tokens, and see who wins for that workload—with batch pricing, delta vs cheapest, and exportable receipts.
Select · price · decide
Catalog leader: text-embedding-3-small
89models in current filters
Tick models in the table, set tokens below, and compare live costs — CentsPerToken’s catalog pricing desk
Models shown89
Cheapest list rate$0.02 / 1M in
Selected0 / 8
Tokens / request2,500
Requests1
Scenario spread—
Scenario
Token workload
Enter the same token shape you would paste into a pricing calculator. Selected models update live.
Live compare
Nothing selected yet
Check up to 8 text models in the table, or auto-pick the cheapest visible ones.
Tip: filter to text, sort by scenario cost, then select models — or use “Pick cheapest visible” to jump straight to a shortlist.
Catalog
List prices & select to compare
Browse like a price desk. Tick comparable text models to price your scenario above.
| Compare | Provider | Modality | ||||||
|---|---|---|---|---|---|---|---|---|
| — | Whisper whisper-1 | OpenAI | audio | — | — | — | — | — |
| — | GPT-4o Transcribe gpt-4o-transcribe | OpenAI | audio | — | — | — | — | — |
| — | Gemini speech-to-text (approx) gemini-speech-to-text | audio | — | — | — | — | — | |
| — | text-embedding-3-small text-embedding-3-small | OpenAI | embedding | $0.02 | — | — | — | — |
| — | DALL·E 3 Standard dall-e-3 | OpenAI | image | — | — | — | — | — |
| — | Imagen 3 imagen-3 | image | — | — | — | — | — | |
| — | Gemini video understanding (approx) gemini-video-understanding | video | — | — | — | — | — | |
| — | GPT Image 1 gpt-image-1 | OpenAI | image | — | — | — | — | — |
| GPT-5 nano gpt-5-nano | OpenAI | text | $0.05 | $0.40 | $0.005 | $0.0003 | 400,000 | |
| — | Flux Pro flux-pro | Black Forest Labs | image | — | — | — | — | — |
| — | GPT-4o video understanding (approx) gpt-4o-video-understanding | OpenAI | video | — | — | — | — | — |
| GPT-5.4 nano gpt-5-4-nano | OpenAI | text | $0.05 | $0.40 | $0.005 | $0.0003 | 272,000 | |
| Gemini 2.0 Flash-Lite gemini-2-0-flash-lite | text | $0.075 | $0.30 | $0.0188 | $0.0003 | 1,000,000 | ||
| — | DALL·E 3 HD dall-e-3-hd | OpenAI | image | — | — | — | — | — |
| Gemini 2.5 Flash-Lite gemini-2-5-flash-lite | text | $0.10 | $0.40 | $0.025 | $0.0004 | 1,000,000 | ||
| GPT-4.1 nano gpt-4-1-nano | OpenAI | text | $0.10 | $0.40 | $0.025 | $0.0004 | 1,047,576 | |
| — | text-embedding-ada-002 text-embedding-ada-002 | OpenAI | embedding | $0.10 | — | — | — | — |
| Gemini 3.1 Flash-Lite gemini-3-1-flash-lite | text | $0.10 | $0.40 | $0.01 | $0.0004 | 1,000,000 | ||
| Gemini 2.0 Flash gemini-2-0-flash | text | $0.10 | $0.40 | $0.025 | $0.0004 | 1,000,000 | ||
| — | text-embedding-3-large text-embedding-3-large | OpenAI | embedding | $0.13 | — | — | — | — |
| DeepSeek V4 Flash deepseek-v4-flash | DeepSeek | text | $0.14 | $0.28 | $0.0028 | $0.00042 | 1,000,000 | |
| GPT-4o mini gpt-4o-mini | OpenAI | text | $0.15 | $0.60 | $0.075 | $0.0006 | 128,000 | |
| — | gemini-embedding-001 gemini-embedding-001 | embedding | $0.15 | — | — | — | — | |
| Grok 4 Fast grok-4-fast | xAI (Grok) | text | $0.20 | $0.50 | $0.05 | $0.00065 | 2,000,000 | |
| Grok 4.1 Fast Reasoning grok-4-1-fast-reasoning | xAI (Grok) | text | $0.20 | $0.50 | $0.05 | $0.00065 | 2,000,000 | |
| Grok 4.1 Fast Non-Reasoning grok-4-1-fast-non-reasoning | xAI (Grok) | text | $0.20 | $0.50 | $0.05 | $0.00065 | 2,000,000 | |
| Grok Code Fast 1 grok-code-fast-1 | xAI (Grok) | text | $0.20 | $1.50 | $0.05 | $0.00115 | 256,000 | |
| GPT-5 mini gpt-5-mini | OpenAI | text | $0.25 | $2.00 | $0.025 | $0.0015 | 400,000 | |
| GPT-5.4 mini gpt-5-4-mini | OpenAI | text | $0.25 | $2.00 | $0.025 | $0.0015 | 272,000 | |
| Claude Haiku 3 claude-haiku-3 | Anthropic | text | $0.25 | $1.25 | $0.03 | $0.001125 | 200,000 | |
| DeepSeek Chat (V3.2) deepseek-chat | DeepSeek | text | $0.28 | $0.42 | $0.028 | $0.00077 | 128,000 | |
| DeepSeek Reasoner (V3.2) deepseek-reasoner | DeepSeek | text | $0.28 | $0.42 | $0.028 | $0.00077 | 128,000 | |
| DeepSeek V3.2 Speciale deepseek-v3-2-speciale | DeepSeek | text | $0.28 | $0.42 | $0.028 | $0.00077 | 128,000 | |
| Gemini 2.5 Flash gemini-2-5-flash | text | $0.30 | $2.50 | $0.075 | $0.00185 | 1,000,000 | ||
| Grok 3 mini grok-3-mini | xAI (Grok) | text | $0.30 | $0.50 | $0.075 | $0.00085 | 131,072 | |
| Gemini 2.5 Flash (fine-tune planning) gemini-2-5-flash-ft | text | $0.30 | $2.50 | — | $0.00185 | 1,000,000 | ||
| — | Veo-style generation (planning) veo-planning | video | — | — | — | — | — | |
| GPT-4.1 mini gpt-4-1-mini | OpenAI | text | $0.40 | $1.60 | $0.10 | $0.0016 | 1,047,576 | |
| — | Sora-style generation (planning) sora-planning | OpenAI | video | — | — | — | — | — |
| GPT-3.5 Turbo gpt-3-5-turbo | OpenAI | text | $0.50 | $1.50 | — | $0.00175 | 16,385 | |
| Gemini 3 Flash Preview gemini-3-flash | text | $0.50 | $3.00 | $0.05 | $0.0025 | 1,000,000 | ||
| DeepSeek V4 Pro deepseek-v4-pro | DeepSeek | text | $0.55 | $2.19 | $0.11 | $0.002195 | 1,000,000 | |
| Claude Haiku 3.5 claude-haiku-3-5 | Anthropic | text | $0.80 | $4.00 | $0.08 | $0.0036 | 200,000 | |
| Claude Haiku 3.5 (fine-tune planning) claude-haiku-3-5-ft | Anthropic | text | $0.80 | $4.00 | $0.08 | $0.0036 | 200,000 | |
| Claude Haiku 4.5 claude-haiku-4-5 | Anthropic | text | $1.00 | $5.00 | $0.10 | $0.0045 | 200,000 | |
| Sonar sonar | Perplexity | text | $1.00 | $1.00 | — | $0.0025 | 128,000 | |
| o4-mini o4-mini | OpenAI | text | $1.10 | $4.40 | $0.275 | $0.0044 | 200,000 | |
| o1-mini o1-mini | OpenAI | text | $1.10 | $4.40 | — | $0.0044 | 128,000 | |
| o3-mini o3-mini | OpenAI | text | $1.10 | $4.40 | $0.55 | $0.0044 | 200,000 | |
| GPT-5 gpt-5 | OpenAI | text | $1.25 | $10.00 | $0.125 | $0.0075 | 400,000 | |
| Gemini 2.5 Pro gemini-2-5-pro | text | $1.25 | $10.00 | $0.315 | $0.0075 | 1,000,000 | ||
| GPT-5.1 gpt-5-1 | OpenAI | text | $1.25 | $10.00 | $0.125 | $0.0075 | 400,000 | |
| Grok 4.3 grok-4-3 | xAI (Grok) | text | $1.25 | $2.50 | — | $0.00375 | 1,000,000 | |
| Gemini 3.5 Flash gemini-3-5-flash | text | $1.50 | $9.00 | $0.15 | $0.0075 | 1,000,000 | ||
| GPT-5.2 gpt-5-2 | OpenAI | text | $1.75 | $14.00 | $0.175 | $0.0105 | 400,000 | |
| Sonar Reasoning Pro sonar-reasoning-pro | Perplexity | text | $2.00 | $8.00 | — | $0.008 | 128,000 | |
| Sonar Deep Research sonar-deep-research | Perplexity | text | $2.00 | $8.00 | — | $0.008 | 128,000 | |
| GPT-4.1 gpt-4-1 | OpenAI | text | $2.00 | $8.00 | $0.50 | $0.008 | 1,047,576 | |
| Gemini 3.1 Pro Preview gemini-3-1-pro | text | $2.00 | $12.00 | $0.20 | $0.01 | 1,000,000 | ||
| Grok 4.20 grok-4-20 | xAI (Grok) | text | $2.00 | $10.00 | $0.50 | $0.009 | 256,000 | |
| Grok 2 Vision grok-2-vision | xAI (Grok) | text | $2.00 | $10.00 | — | $0.009 | 32,768 | |
| GPT-4o gpt-4o | OpenAI | text | $2.50 | $10.00 | $1.25 | $0.01 | 128,000 | |
| GPT-5.4 gpt-5-4 | OpenAI | text | $2.50 | $15.00 | $0.25 | $0.0125 | 272,000 | |
| Claude Sonnet 4 claude-sonnet-4 | Anthropic | text | $3.00 | $15.00 | $0.30 | $0.0135 | 200,000 | |
| Claude Sonnet 4.5 claude-sonnet-4-5 | Anthropic | text | $3.00 | $15.00 | $0.30 | $0.0135 | 200,000 | |
| Sonar Pro sonar-pro | Perplexity | text | $3.00 | $15.00 | — | $0.0135 | 200,000 | |
| Grok 3 grok-3 | xAI (Grok) | text | $3.00 | $15.00 | $0.75 | $0.0135 | 131,072 | |
| Grok 4 grok-4 | xAI (Grok) | text | $3.00 | $15.00 | $0.75 | $0.0135 | 256,000 | |
| Claude Sonnet 4.6 claude-sonnet-4-6 | Anthropic | text | $3.00 | $15.00 | $0.30 | $0.0135 | 200,000 | |
| Claude Sonnet 3.7 claude-sonnet-3-7 | Anthropic | text | $3.00 | $15.00 | $0.30 | $0.0135 | 200,000 | |
| Claude Opus 4.5 claude-opus-4-5 | Anthropic | text | $5.00 | $25.00 | $0.50 | $0.0225 | 200,000 | |
| GPT-5.5 gpt-5-5 | OpenAI | text | $5.00 | $30.00 | $0.50 | $0.025 | 272,000 | |
| Claude Opus 4.8 claude-opus-4-8 | Anthropic | text | $5.00 | $25.00 | $0.50 | $0.0225 | 200,000 | |
| Claude Opus 4.7 claude-opus-4-7 | Anthropic | text | $5.00 | $25.00 | $0.50 | $0.0225 | 200,000 | |
| Claude Opus 4.6 claude-opus-4-6 | Anthropic | text | $5.00 | $25.00 | $0.50 | $0.0225 | 200,000 | |
| o3 o3 | OpenAI | text | $10.00 | $40.00 | $2.50 | $0.04 | 200,000 | |
| GPT-4 Turbo gpt-4-turbo | OpenAI | text | $10.00 | $30.00 | — | $0.035 | 128,000 | |
| Claude Opus 4 claude-opus-4 | Anthropic | text | $15.00 | $75.00 | $1.50 | $0.0675 | 200,000 | |
| — | TTS-1 tts-1 | OpenAI | audio | — | — | — | — | — |
| GPT-5.2 Pro gpt-5-2-pro | OpenAI | text | $15.00 | $120.00 | $1.50 | $0.09 | 400,000 | |
| GPT-5 Pro gpt-5-pro | OpenAI | text | $15.00 | $120.00 | $1.50 | $0.09 | 400,000 | |
| o1 o1 | OpenAI | text | $15.00 | $60.00 | $7.50 | $0.06 | 200,000 | |
| Claude Opus 4.1 claude-opus-4-1 | Anthropic | text | $15.00 | $75.00 | $1.50 | $0.0675 | 200,000 | |
| Claude Opus 3 claude-opus-3 | Anthropic | text | $15.00 | $75.00 | $1.50 | $0.0675 | 200,000 | |
| o3-pro o3-pro | OpenAI | text | $20.00 | $80.00 | $5.00 | $0.08 | 200,000 | |
| — | TTS-1 HD tts-1-hd | OpenAI | audio | — | — | — | — | — |
| GPT-5.5 Pro gpt-5-5-pro | OpenAI | text | $30.00 | $180.00 | $3.00 | $0.15 | 272,000 | |
| GPT-5.4 Pro gpt-5-4-pro | OpenAI | text | $30.00 | $180.00 | $3.00 | $0.15 | 272,000 | |
| o1-pro o1-pro | OpenAI | text | $150.00 | $600.00 | — | $0.60 | 200,000 |
Scenario costs use CentsPerToken list prices for planning only—not negotiated rates. Different models tokenize differently, so token counts are not perfectly transferable. Showing 89 models. Updated 2026-07-31. Approximate static list prices for planning only. Always verify against each provider’s official pricing page before production budgeting.
Guide
How to get value from this calculator
Browse list prices across providers, then select models and price a real token scenario (input, cached input, output, requests) in one desk. Unlike a static rate sheet, CentsPerToken ranks your shortlist by scenario cost, shows delta vs the winner, and exports a decision receipt—while still supporting image, audio, and embedding unit rates in the catalog.
Why it matters
List prices alone do not answer “what does my prompt cost?” Scenario pricing turns a rate table into a decision tool—especially when output tokens are 3–5× more expensive than input on many models.
How to use it
- Filter or search the catalog to the models you care about.
- Enter input, cached input, output tokens, and request count—or pick a scenario preset.
- Tick up to eight text models (or use “Pick cheapest visible”).
- Read the live ranked compare panel for winner, spread, and % delta.
- Toggle batch pricing if your workload can wait for batch APIs.
- Copy the decision receipt or export CSV, then verify winners on each provider’s official pricing page.
Planning tips
- Sort by scenario cost after selecting models—input $/1M alone misleads output-heavy apps.
- Use cached input tokens when you have a stable system prompt or RAG prefix.
- “Pick cheapest visible” is a shortlist generator, not a production recommendation.
- Enterprise discounts are not in this table—re-run with your negotiated rates mentally or in a custom budget tool.
- For deep multi-provider routing with paste-prompt estimates, continue to the API cost calculator.
- Different tokenizers mean the same English text is not the same token count across providers.
Frequently asked questions
How often is the price table updated?
CentsPerToken refreshes list prices periodically, but there may be lag versus provider announcements. Always confirm current rates officially before production decisions.
Why do some models show very different input vs output prices?
Providers set output rates higher because generation is more compute-intensive. The ratio varies by model generation and provider strategy.
Does the table include batch or cached pricing?
The table shows standard list prices. Batch discounts and prompt cache rates require the batch vs realtime and cache savings tools respectively.
Can I use this for invoice reconciliation?
This is a planning reference, not an billing system. Use the usage CSV analyzer with your export data for invoice-style reconciliation.