2026 LLM Token Cost Ledger & Agent Config Workbench
Compare 30-day API spend across GPT-6.1, Claude 5.5, Gemini 4, DeepSeek V4.1, Grok 4.7, and Qwen 3.8—100% locally in your browser.
| Model (2026 Release) | Provider & date | Input / cached ($/1M) | Output ($/1M) | Per 1K calls | 30-day spend |
|---|---|---|---|---|---|
| Gemini 4 Hydrogen | Google 2026-11-12 | $1.5 / $0.075 | $7.5 | $9.1500 | |
| Gemini 4 Carbon | Google 2026-11-10 | $2 / $0.1 | $10 | $12.2000 | |
| Claude Fable 5.2 | Anthropic 2026-11-09 | $8 / $0.2 | $40 | $48.4000 | |
| Muse Ember | Meta 2026-11-09 | $0.1 / $0.004 | $0.5 | $0.6080 | |
| Gemini Omni 1.2 Flash | Google 2026-11-05 | $1.5 / $0.375 | $9 | $10.9500 | |
| Gemini Omni 1.1 Flash | Google 2026-11-04 | $1.5 / $0.375 | $9 | $10.9500 | |
| GLM 5.3 FlashX Legal | Z.ai / GLM 2026-11-02 | $0.37 / $0.075 | $1.25 | $1.8900 | |
| GPT-6.1 Terra Pro | OpenAI 2026-10-26 | $1 / $0.05 | $5 | $6.1000 | |
| GPT-6.1 Terra | OpenAI 2026-10-26 | $1 / $0.05 | $5 | $6.1000 | |
| Kimi K3.1 Pro | Moonshot 2026-10-23 | $2 / $0.2 | $10 | $12.4000 | |
| MiniMax M3.1 Flash | MiniMax 2026-10-19 | $0.15 / $0.015 | $0.6 | $0.8100 | |
| DeepSeek V4.1 Flash 1017 | DeepSeek 2026-10-17 | $0.038 / $0.01 | $0.5 | $0.4960 |
Client-Side FinOps & Agent Configuration Suite
Part of CodeAP Studio alongside ToMarkdown, DevToken pairs an automated 12-hour verified pricing snapshot (2026-11-13) with zero-upload browser utilities. See our Methodology for token heuristic details and caching formulas.
Frequently Asked Questions
How does DevToken keep its LLM pricing and model release data current?
A scheduled GitHub Actions workflow runs the DevToken 12-hour verified pricing pipeline (sync-devtoken-models.yml),
curating 60 active 2026 models across 472+ tracked endpoints into a zero-latency static snapshot.
Are my prompts, MCP credentials, or JSON schemas sent to any backend?
No. Every BPE token estimate, MCP configuration validation, llms.txt build, and strict JSON Schema or Zod compilation
executes 100% locally inside your browser with zero server round-trips.
How can I convert PDF or HTML files before estimating RAG token costs?
Open ToMarkdown to convert PDF, Office, HTML, or CSV files into clean GFM Markdown, then click Estimate LLM Cost to pass the exact token count directly into this ledger.