Make yours

Themes, providers, workflows, and tools, installed with one command.

/
Token Usage & Limits29

Token Usage & Limits 29

Understand or control token, context-window, and provider-quota use.

BurnSee what every agent thread costs in dollars while it runs, per prompt and across all your threads. A chip beside the context circle shows the thread's total and what the last prompt added; click it for where the money went, a chart of every model call and the total with child threads. The Burn tab in the side panel lists the cost of each prompt, and a Burn page in bb's navigation charts spend across threads by day, project, model and harness. Budgets notify you at 80% and 100% of a thread's budget, and stop a run only if you turn that on. Works with pi, OpenCode, Codex, Claude Code and any harness that reports usage to bb; where a harness records no cost, Burn estimates it from list prices and marks it with ~.Token Usage & LimitsKalindu De Costa57
Context KatamariShows how full each thread's context window is as a katamari that grows with every turn and pops when bb compacts. A floating window gives the open thread a 3D toy-box world whose ball size tracks tokens used against the auto-compact point. The HUD shows tokens used, context left, a bar for each recent turn, and a count of compactions. The cousin rolls while the thread works, drops coins past 50k tokens, and gets dizzier after each compaction. Everything runs inside bb, with synthesized sound you can mute.Token Usage & LimitsbrsblNew
Higgsfield CreditsShows how many Higgsfield credits you have, when they reset, and how many will expire unused at your current pace. A burn forecast charts this cycle's balance and projects it to the reset date, so you see whether you run out early or leave credits behind. A per-model table counts generations, refunded failures, credits, and cost per generation, tagged image, video, audio, or text. A daily chart stacks spend by model or media type for this cycle, 7, 30, or 90 days. Requires the Higgsfield CLI and a Higgsfield account; sign-in uses Higgsfield's own browser login.Token Usage & LimitsVladPoletNew
Model PricingCompare what every AI model costs per million tokens, across Anthropic, OpenAI, Fireworks AI, Ollama Cloud and OpenRouter, in one table. A Relative column ranks each model against the cheapest one on screen, weighted by an input/output blend you choose. Each row also gives the context window, the longest reply, and whether the model can reason, call tools or return structured output. Sort by any column, search by name, hide a provider with its legend dot, and open any model's own page from its row. Prices come from the public models.dev catalog and OpenRouter's model list and are kept for twelve hours; no account or key is needed, but the machine running the bb server must reach both sites.Token Usage & LimitsGustavo AmbrozioNew
Pool UsageSee how much of each provider's capacity is in use and when it frees up, right in the sidebar footer. Each provider gets its logo and one percentage, with accounts weighed by their plans, so a Claude Max 20x seat counts twenty times a Pro seat. Clicking a number opens a card with the account count and when the total drops next. When Account Pooler's Claude or Codex accounts hold resets, the card offers the one that frees the most capacity and spends it only after you confirm. Covers Account Pooler's accounts and any provider that reports usage to BB, such as Cursor; Account Pooler is optional.Token Usage & LimitsYurii Rebryk37
TokenomicsSee what your threads cost in tokens and time, and whether your subscription's session and weekly limits will last until they reset. Each thread's header shows a sparkline of its token use that opens a summary of which of your messages used the most tokens and time. bb's title bar shows how much of your session and week you have used, in amber or red when either is on pace to run out before it resets. Once a thread's context passes a size you set, a meter above the composer offers to compact it. A Tokenomics page charts tokens by hour or day, shows where turn time went, and ranks threads by use. Counts tokens, not dollars, needs bb 0.43 or later, and counts Claude Code subagents only for threads that ran on the same machine as bb.Token Usage & LimitsDaniel BachhuberNew
Turn TPSShow tokens per second on every completed turn, so a slow answer is visible without opening a log. Each finished turn gets a badge under the assistant message with output tokens, wall-clock duration, and the resulting rate, and the turn header carries the same figure in short form. Reads usage the provider reports where BB has it, and the local OMP session file for ACP and OMP threads, which report none. The rate divides output tokens by the full turn duration, so tool calls and provider waits count against it; it is not a model decode benchmark. Needs a completed turn, and the ACP and OMP path needs that thread's OMP session file on the machine running BB.Token Usage & LimitsDaniel WalkerNew
Usage PaceTells you if your current rate lasts until each quota window resets. In bb's usage card, each bar gets a tick at even pace, and red shows how far ahead of pace you are. Hover or tap a row to see the difference, when the quota runs out, and how long you will be without quota before the reset. Adds a Grok Build tab, shows a failed provider's error on its own tab only, and hides tabs you do not use, such as Cursor. The Grok tab reads your grok login and asks xAI's billing service for the credit window; everything else uses the data bb's card already has.Token Usage & LimitsCharles LeeNew