# TokenPricing Review Free LLM API price comparison tool covering 627 models and 236 providers, with input, output, and cached token rates side by side. ## Editor's Score: 70/100 Capability scores 18 because the breadth of 627 models and 236 providers is genuinely strong, but the absence of a public API, quality-adjusted scoring, and any non-standard pricing coverage are real gaps that named alternatives like TokenPrice.cloud partially fill. Ease of use scores 22 because the no-login, no-friction design and clear table layout are about as low-barrier as a web tool gets, with only minor complexity in reading expanded provider rows. Value scores 25 because the tool is entirely free with no paywalled features, which is the maximum possible value for a reference tool of this kind. Delivery scores 5 because the dedicated social search could not be completed due to tool quota exhaustion, meaning no user sentiment, no community signal, and no independent verification of data freshness or accuracy could be gathered; the human editor must verify these before the score can be raised with confidence. ## Pros - Covers 627 models and 236 providers, with per-provider price spread visible at a glance for the same model, making arbitrage between providers immediately obvious. - Sourcing discipline is strong: each price links to its published source, missing prices stay blank instead of defaulting to zero, and outliers are held until verified. - Completely free with no login, no rate limits on browsing, and no paid tier obscuring any data. ## Cons - No programmatic access: there is no public API for TokenPricing's own dataset, so developers cannot pull live pricing into their own cost-monitoring dashboards or CI pipelines without scraping. - No quality-to-price scoring: the tool shows raw token costs but offers no benchmark-weighted value signal, so a model that is cheap but slow or low-quality looks identical to a cheap, high-quality one in the table. - Coverage of non-standard pricing structures is intentionally excluded, meaning batch discounts, token-tier rates, and region-specific prices, which can represent significant real-world savings, are absent from every comparison. ## Pricing Free. Pricing reflects what we saw at the time of review (2026-10-21). Confirm current pricing on the official site. Category: AI Coding & Dev Tools (https://aitoolseekers.com/category/ai-coding-dev-tools) Official site: https://tokenpricing.co/ Last verified by a human: 2026-10-21 Reviewed by a human at AI Tool Seekers, a hand-reviewed AI tools directory. Every listed tool is verified by a person before publishing. Scores are editorial opinions from a fixed rubric (capability, ease of use, value, delivery on promise) and cannot be bought; there is no pay-to-rank. Directory: https://aitoolseekers.com This review: https://aitoolseekers.com/tools/tokenpricing Methodology: https://aitoolseekers.com/how-verification-works ## Verdict TokenPricing is the go-to reference for developers who need to compare token costs across a large number of providers before committing to an API. Its coverage of 627 models and 236 providers is wider than most comparable free tools, but it is a read-only reference with no programmatic API of its own, which limits its usefulness for teams wanting to automate cost monitoring. If you need benchmark-weighted value scores or quality-to-price ratios alongside raw pricing, a tool like TokenPrice.cloud edges it out on that dimension. Best for: Developers comparing LLM API costs across many providers before choosing or switching models. Not for: Teams needing automated cost alerts, webhook integrations, or quality-adjusted value scoring built into the same tool. ## Overview TokenPricing is a free, independent reference site that tracks LLM API token costs across 627 models and 236 providers, displaying input, output, and cached token prices in USD per 1 million tokens. The core value is breadth: a single table lets a developer compare what Relace, DeepInfra, Novita, Amazon Bedrock, and dozens of other providers charge for the same model, without having to visit each provider's pricing page separately. The site is built around a few distinct tools: a model pricing table, a provider comparison view, and a cost calculator. The pricing table shows one base-rate row per model by default, with expandable rows for every additional provider offer. Missing prices are left blank rather than filled with zeros, which is a meaningful design choice that prevents misleading cost comparisons. Batch pricing, region-specific rates, VIP-group offers, and token-tier discounts are deliberately excluded so the table reflects standard, publicly available rates only. What separates TokenPricing from a simple spreadsheet is the sourcing discipline: each offer links to the page where the price was published, and outlier prices are held out of summaries until verified. As of the live page, models like Claude Opus 5.5 show consistent $4.00 input and $20.00 output pricing per 1M tokens across five providers, while DeepSeek V4.1 Flash shows meaningful spread, from $0.05 input at Relace to $0.135 at AtlasCloud, which is exactly the kind of arbitrage signal a cost-conscious developer needs. The tool is entirely web-based, requires no account, and appears to be maintained by a small independent team. There is no paid tier and no upsell path visible anywhere on the site. ## Who should use it Solo developers and small engineering teams who are actively evaluating which LLM provider to call for a new project will get the most out of TokenPricing. The ability to see, for example, that DeepSeek V4.1 Flash input pricing ranges from $0.05 to $0.135 per 1M tokens depending on provider is immediately actionable when you are writing infrastructure code and need to pick a vendor. Product managers and AI researchers who want a quick sanity check on whether a vendor's quoted price is competitive will also find it useful. However, if you need quality-adjusted value scores that weight price against benchmark performance, TokenPrice.cloud adds that dimension. If you are building an agent pipeline and need to route between providers programmatically based on live cost data, Qveris, which is already in this directory, is a better fit since it connects AI agents to APIs through a single protocol rather than just displaying prices. ## FAQ Q: Does TokenPricing require an account or login? A: No. The site is fully accessible without signup or login, and all features including the pricing calculator are free to use. Q: How many models and providers does TokenPricing cover? A: As of October 2026, the live page lists 627 models and 236 providers. The table is updated as new models and provider offers are published. Q: Does TokenPricing include batch or discounted pricing tiers? A: No. Only active, standard base token rates are included. Batch pricing, region-specific rates, VIP-group offers, and token-tier discounts are deliberately excluded to keep comparisons consistent. Q: Is there a cost calculator, and what does it do? A: Yes. The calculator lets you enter token volumes and see an estimated API bill for a chosen model and provider, using the same per-1M-token figures shown in the main table. Q: Does TokenPricing have its own API so I can pull pricing data into my app? A: No public API for TokenPricing's own dataset is listed on the site. A separate GitHub project called tokenpricing (by Atena-IT) offers a Python SDK and CLI that syncs from OpenRouter and LiteLLM every six hours, but that is a distinct, unaffiliated project. Q: How does TokenPricing handle prices that have not been published by a provider? A: Missing prices are left blank and shown as unknown rather than defaulting to zero, and price outliers are excluded from summaries until verified.