Claude Fable 5.1 Pricing on Quanta: One Balance for 120 RPM

01 · Should a growing team buy seat subscriptions or use one metered Quanta balance?

For a team whose usage rises week to week, seat subscriptions create a recurring obligation that may sit unused or run short. Quanta takes a different approach: there is one prepaid balance, metered on the real tokens each call consumes, with no monthly fee and no minimum spend. That makes Quanta a practical fit for growing teams that want to pay for actual work rather than commit to a fixed arrangement in advance.

The point is not simply flexibility. It is visibility. Quanta publishes the limits that affect heavy, sustained workloads, and the dashboard lists the tokens and charge of every call. Input, output, cache writes and cache reads are priced separately, so a team can see where its usage actually goes. New accounts also receive $0.50 in trial credit, which is enough to run a first request and confirm that the setup behaves as expected.

There is one important boundary to keep in mind. Quanta is not a seat-based workspace. It is an API relay, and access is organised around keys, balances and metered calls. If your team wants to route sustained model traffic through one account, with limits and invoices stated plainly, Quanta is designed for that shape of work.

02 · What does Claude Fable 5.1 cost on Quanta, and which clients can use it?

Claude Fable 5.1 is priced per 1M tokens on Quanta at $4.38 for input, $21.88 for output, $5.39 for cache writes and $0.3333 for cache reads. These are approximate USD figures; charges are made in CNY. Because output tokens carry a different rate from input tokens, a workload that produces long responses will cost more than one that mostly reads context, and Quanta shows both lines per call so the difference is visible rather than inferred.

Client pairing matters here. Claude Fable 5.1 works only from the official Claude Code and Claude Desktop clients, with Claude Code 2.1.285 or newer. It does not work through Codex, Cursor, Cherry Studio, Cline, OpenCode, an SDK or cURL. Teams that standardise on Claude Code or Claude Desktop can use Claude Fable 5.1 directly; teams running other tools should choose from the GPT, DeepSeek and other non-Claude, non-Gemini models available on Quanta.

• › Claude Code and Claude Desktop: Claude Fable 5.1 and the other listed Claude models.
• › Gemini CLI: Gemini models over the native protocol.
• › Codex, Cursor, Cherry Studio, Cline, OpenCode, OpenAI SDKs and cURL: GPT, DeepSeek and the other listed non-Claude, non-Gemini models.

One key and one balance work across the OpenAI-compatible (/v1), Anthropic native and Gemini native protocols. That matters for a team that does not want every tool to point at a different vendor relationship.

03 · How does one metered balance hold up under heavy, sustained workloads?

Quanta publishes and enforces the limits that a busy team will actually meet. Each key is limited to 120 requests per minute, and an account may have 8 requests in flight. Request bodies are capped at 25 MB, and an account can hold 20 live keys. Streams have no total-time cap and end only after 300 seconds of upstream silence, which suits long-running generation tasks that would otherwise be interrupted by a short timeout.

These are the operating conditions of the service, not a hidden tier structure. A team can plan its concurrency and key allocation against numbers that are stated in advance, and it can see each call's token counts and charge in the dashboard as the workload runs.

Top-ups are available at $1.39, $6.94, $13.89, $27.78, $69.44 and $138.89. Because there is no minimum spend, a team can start with the smallest amount and scale up in line with measured usage. Unused balance can be refunded within 15 days of the top-up, which gives a growing team room to adjust its pace without treating every top-up as a permanent commitment.

For teams that bring colleagues in, referrals also carry real value: 19% of a referred account's top-ups, and 5% one level further. Rewards can be withdrawn from $27.78 or moved into the balance.

04 · What does a typical month look like on Quanta?

The following are examples, not quotes or promises. Each assumes a 3:1 input to output ratio with no cache hits, and shows the approximate cost of 10 million and 100 million tokens in a month.

• › Claude Opus 5.5: 10 million tokens a month ≈ $34.58; 100 million ≈ $345.83.
• › Claude Sonnet 5.5: 10 million tokens a month ≈ $25.07; 100 million ≈ $250.69.
• › GPT-6.1 Sol: 10 million tokens a month ≈ $4.10; 100 million ≈ $40.97.
• › Gemini 2.5 Flash: 10 million tokens a month ≈ $6.49; 100 million ≈ $64.93.

Say your team routes a nightly batch through Claude Sonnet 5.5 and interactive requests through GPT-6.1 Sol. As an example, 100 million tokens a month on Claude Sonnet 5.5 at ≈ $250.69, plus 100 million tokens a month on GPT-6.1 Sol at ≈ $40.97, comes to roughly $291.66 in metered charges for that illustration. The same traffic on Quanta appears line by line in the dashboard, so a team can shift volume between models based on measured cost and observed quality.

Cache pricing is part of that calculation. Claude Fable 5.1 charges $0.3333 per 1M cache read tokens and $5.39 per 1M cache write tokens, while models such as GPT-6.1 Sol list no cache write rate and $0.0403 per 1M cache read tokens. For repeated context, those figures are worth reading before a workload is scaled.

05 · How do teams handle invoices, and what does Quanta keep?

Company billing is handled through bank transfer from a company account. Invoices and advance quotes are arranged through the Contact page or the Messages page of the dashboard, so procurement questions stay attached to the account rather than in a separate thread. There is no monthly fee to justify and no minimum spend to reach before usage begins counting.

Data handling is stated just as directly. The text users send is kept for 30 days for abuse handling and billing disputes, then deleted. Model answers are not stored. For teams handling sensitive material, that is a concrete retention window to plan around rather than a vague assurance.

Quanta also supports the full range of listed models behind the same balance: Claude Haiku 4.5 at $0.4167 input and $1.97 output, Claude Opus 5 at $1.97 and $9.83, Gemini 3.1 Pro at $1.47 and $8.56, Gemini 3.8 Flash at $1.35 and $6.10, Gemini 4 Argon at $1.72 and $8.28, DeepSeek V4 Pro at $1.44 and $4.32, Kimi K3 at $1.31 and $6.54, Qwen3.8 Max at $0.8611 and $2.58, Hunyuan 4 at $0.4167 and $1.32, Grok 4.7 at $0.625 and $1.88, and GLM-5.3 at $0.6528 and $1.97, all per 1M tokens.

06 · Why open a Quanta account for a growing team?

Because the alternative to a seat subscription is not another fixed commitment. On Quanta, a growing team pays as it goes, sees the tokens and charge for every call, and knows the concurrency, rate limits and request sizes it is working within before traffic builds up.

Start with the $0.50 trial credit, point Claude Code or Claude Desktop at Quanta to try Claude Fable 5.1, and check the dashboard. If the per-call breakdown matches expectations, top up the amount that fits the current month. From there, Quanta keeps the numbers visible, the balance flexible, and the invoice path straightforward.