GPT-5.6 Sol on BestRelay: $0.4167 input pricing for API relay

BestRelay sends each request to the model the caller named

BestRelay is an independent API relay, and its central promise is plain: the model you name is the model that answers. Each request is forwarded to the model provider under the model name the caller wrote, so nothing is quietly exchanged for a cheaper tier behind the scenes. That matters when an agent is doing real work over a month, because the price on the card and the model doing the job should agree. BestRelay keeps a curated line-up of 20 public models, each listed under its own name with four prices: input, output, cache write and cache read. When a provider retires a model id, BestRelay delists it rather than leaving a stale entry in place.

The practical effect is that you can plan an agent's running cost against a named model and a stated rate, instead of guessing at what a relay might substitute. The line-up includes GPT-5.6 Sol, GPT-6.1 Sol, GPT-6 Astra, Claude Fable 5, Claude Fable 5.1, Claude Haiku 4.5, Claude Opus 5, Claude Opus 5.5, Claude Sonnet 5, Claude Sonnet 5.5, DeepSeek V4 Pro, Gemini 2.5 Flash, Gemini 3.1 Pro, Gemini 3.8 Flash, Gemini 4 Argon, Kimi K3, Qwen3.8 Max, Grok 4.6, Grok 4.7 and GLM-5.3.

GPT-5.6 Sol costs $0.4167 per 1,000,000 input tokens

For example, GPT-5.6 Sol on BestRelay is priced at $0.4167 per 1,000,000 input tokens, $2.29 per 1,000,000 output tokens and $0.0333 per 1,000,000 cache read tokens. The official list price for the same model is $4 input, $20 output and $0.4 cache read. That gap is worth looking at over a month of agent work, because agent loops tend to send a good deal of context in and read a good deal of it back from cache.

Take the example given for a day of 500,000 input and 100,000 output tokens: on BestRelay this costs $0.44, and 22 such days cost $9.62. With 350,000 of the input read from cache, the day costs $0.30. At the official list price, the same day is $4.00, the 22 days are $88.00 and the cached day is $2.74, which puts the difference at $3.56 for the day and $78.38 across the 22 days. The example is only arithmetic, of course, but it shows the shape of the saving clearly.

Claude Sonnet 5.5 and GPT-6.1 Sol for lighter days

The same exercise on Claude Sonnet 5.5 gives a day of 500,000 input and 100,000 output tokens at $1.40 on BestRelay, with 22 such days at $30.71, and a cached day at $0.94. At the official list price the day is $2.00, the 22 days $44.00 and the cached day $1.33, so the difference is $0.60 on the day and $13.29 across the 22 days. Claude Sonnet 5.5 is priced at $1.39 input, $7.01 output and $0.0944 cache read per 1,000,000 tokens on BestRelay.

GPT-6.1 Sol is the lighter entry in the line-up at $0.2083 input, $1.11 output and $0.0431 cache read per 1,000,000 tokens. For example, a day of 500,000 input and 100,000 output tokens costs $0.22 on BestRelay, and 22 such days cost $4.74; with 350,000 of the input read from cache, the day costs $0.16. At the official list price the day is $2.00, the 22 days $44.00 and the cached day $1.33, a difference of $1.78 on the day and $39.26 over the 22 days. The $0.35 of credit that opens a new BestRelay account covers about 1.6 million GPT-6.1 Sol input tokens on its own.

Pay-as-you-go billing with no monthly fee

BestRelay bills on a prepaid pay-as-you-go balance. There is no monthly fee, no subscription and no minimum spend. Each call is charged once, after it ends, from the tokens the provider reports, at the rate card prices. Nothing is held in advance, which means a balance is only reduced by work that has actually run.

Failed calls are handled without ceremony. Requests the provider rejects are recorded at zero, and when the provider cannot be reached, nothing is charged. Top-ups run through Stripe checkout in these amounts: $1.39, $6.94, $13.89, $27.78, $69.44 and $138.89. Unused credit can be refunded on request within 13 days of the top-up, less what was consumed and any payment-processor fees; gift credit and commission cannot be refunded.

There is also a modest referral arrangement: 18% of the top-ups of people who joined through your link, and 4% of the top-ups of the people they invited. Commission can be moved into the balance or withdrawn to Alipay from $20.83. For example, on an invited member's top-up of $13.89, the two commissions are $2.50 and $0.5556.

Setup, clients and the shape of the line-up

Connecting a client to BestRelay needs two settings, the address and a key created under API keys in the dashboard; the details are in the docs. A key carries limits of 120 requests a minute and 8 calls in flight on one account. An account can hold several keys, each can be voided at any time, and every call appears in the usage records with its model, tokens and charge.

Client support follows the models themselves. Claude models are served only to the official Claude Code and Claude Desktop clients, version 2.1.285 or newer. Gemini models use only their native protocol, through Gemini CLI or direct generateContent calls. Tools such as Codex pair with GPT, DeepSeek, Kimi, GLM, Qwen and Grok.

On privacy, the user-written text of requests is kept for 30 days to review abuse and settle billing questions; system prompts and model replies are not stored. Taken together, the arrangement is straightforward: name the model you want, see four prices for it, and pay for the calls that finish. If you would like to try BestRelay, an account opens with $0.35 of credit.