COST MODELLING / VERIFIED SEP 30 2026
Model the cost before it scales.
Compare current models from OpenAI, Anthropic, Google, and xAI. Model your workload and cache reads using standard text API rates in USD per 1M tokens.
Workload inputs
Estimated / month
$748.00
GPT-6.1 Sol
Cache savings
$152.00
vs. fully uncached
Standard text-token estimate. Include billed reasoning tokens in output. Cache reads are modeled; cache writes, storage, tools, search, long-context uplifts, regional premiums, taxes, and discounts are excluded.
PRICE BOARD
Public model rates.
A quick comparison—not a benchmark. Choose models on quality, latency, reliability, and task fit before optimizing token cost.
LIVE CATALOG / OPENROUTER
Pull the latest listed rates.
The endpoint is public, CORS-enabled, and returns per-token prompt, completion, and cache fields. Treat it as an aggregator feed; keep official provider pages as the source of record.
GET https://openrouter.ai/api/v1/models?output_modalities=text&model_authors=openai,anthropic,google,x-ai&sort=pricing-low-to-high
Ready to query the live catalog.
| Provider / model | Input | Cached | Output | Note | Source |
|---|---|---|---|---|---|
| OpenAIGPT-6.1 Sol | $2.00 | $0.10 | $10.00 | Standard; prompts ≤272K tokens | Official |
| OpenAIGPT-6 Astra | $10.00 | $1.00 | $50.00 | Standard; prompts ≤272K tokens | Official |
| OpenAIGPT-6 Luna | $0.10 | $0.01 | $0.50 | Standard; prompts ≤272K tokens | Official |
| OpenAIGPT-5.6 Sol | $4.00 | $0.40 | $20.00 | Promo through at least Nov 21, 2026; ≤272K | Official |
| AnthropicClaude Fable 5.1 | $10.00 | $0.25 | $50.00 | Standard; cache writes billed separately | Official |
| AnthropicClaude Opus 5.5 | $4.00 | $0.20 | $20.00 | Standard; cache writes billed separately | Official |
| AnthropicClaude Sonnet 5.5 | $2.00 | $0.20 | $10.00 | Standard; cache writes billed separately | Official |
| AnthropicClaude Opus 5 | $5.00 | $0.50 | $25.00 | Previous generation; standard rate | Official |
| AnthropicClaude Haiku 4.5 | $1.00 | $0.10 | $5.00 | Standard rate | Official |
| GoogleGemini 3.8 Flash | $0.75 | $0.075 | $3.75 | Promo through Dec 31, 2026 | Official |
| GoogleGemini 3.1 Pro Preview | $2.00 | $0.20 | $12.00 | Prompts ≤200K tokens; cache storage extra | Official |
| GoogleGemini 3.5 Flash-Lite | $0.30 | $0.03 | $2.50 | Standard rate | Official |
| xAIGrok 4.7 | $2.00 | $0.50 | $6.00 | Standard; prompts ≤200K tokens | Official |
OPERATING NOTES
The cheapest token is the one you do not send.
Cache stable system context and reusable documents.
Route routine work to smaller models; escalate on uncertainty.
Cap outputs and stop agent loops when marginal value drops.
Track cost per successful task—not simply cost per token.