tokcalc

The open-source LLM serving capacity planner

Plan your LLM deployment before you rent the GPUs.

tokcalc demo

Switch model → multi-GPU → long-context capacity planner → Build vs Buy → Reference catalog → Live GPU pricing

tokcalc turns your LLM traffic, context length, latency SLOs, cache behavior, and model choice into a defensible serving topology and cost plan — with transparent formulas and cited benchmarks.

It's the engineering-grade pre-deployment decision layer for LLM inference. Not another static "tokens per second" calculator.

Try it now

tokcalc.vercel.app — no signup, no tracking, no paywall.

Pick a model, a GPU, and a workload. Get an instant capacity plan:

Generation speed (tok/s)
Time-to-first-token (TTFT) + inter-token latency (ITL)
VRAM budget with KV-cache sizing
Multi-GPU topology recommendation (Single GPU → TP×2/4/8 → Context Parallel)
Monthly cost + break-even vs API pricing
Live GPU pricing across Azure + AWS + GCP + Vast.ai marketplace (fetched in parallel)
Shareable URL — your config encoded in the URL hash, send to colleagues
Copy as Markdown — paste the full result into GitHub issues / Slack / docs

Why tokcalc?

The market has dozens of "tokens per second" calculators and self-host-vs-API break-even tools (induwara.lk, gigagpu, kickllm, cloudparity, curlscape, profitable.ai). None of them are unified capacity planners — and none show their math.

What tokcalc answers that competitors can't

"Can I serve Qwen 2.5 72B at 128K context on 2× H100 with 20 concurrent users?"

"How many H200s do I need for 1,000 req/min with P95 TTFT < 2s?"

"Does FP8 or AWQ save more money once quality, KV cache, and engine support are included?"

"At what daily volume does an H100 beat GPT-4o pricing?"

"What's the cheapest H100 right now across Azure / AWS / GCP / Vast.ai?"

"What happens to cost and latency if an agent makes 8 model calls, has 3 tool calls, and its context grows by 5K tokens each turn?"

"Would prefix caching, continuous batching, or PD disaggregation save more for this workload?"

Features

Calculator tab

Feature	What it computes
Model fit / VRAM	Will the model + KV cache fit in the GPU's memory?
Throughput	Decode tok/s (per-stream) + aggregate (batched) + prefill tok/s
Latency split	Time-to-first-token (= prefill) + inter-token latency (= decode)
Continuous batching	User-tunable 1.0–4× multiplier (cited 1.5–4× SOSP range)
Reasoning tokens	Hidden reasoning budget added to billed output (o1 / R1 / Claude thinking)
Prompt caching	Self-hosted vLLM APC + Anthropic 5m/1h TTL + OpenAI 50%-off cached tokens
Speculative decoding	User-tunable 1.2–4× boost factor
Multi-GPU TP	1× → 8× tensor parallel with NVLink efficiency factor
Long-context capacity	KV memory + max concurrency + prefill time at 4K → 1M context
Topology recommendation	Single GPU → TP×2 → TP×4 → TP×8 → TP×8 + Context Parallel (RingAttention)
Cost economics	GPU $/hr → $/M output tokens → $/request → monthly cost
Observed benchmark calibration	Paste vLLM/SGLang/TRT-LLM JSON → see formula accuracy verdict (validated / underestimated / overestimated)

Build vs Buy tab

Independent calculator (separate state) that compares:

Self-host: model + GPU + quant + utilization + batch → $/M tokens + monthly cost
API: 13 providers (OpenAI / Anthropic / Gemini / Groq / DeepSeek / Mistral / Together)
Verdict: Self-host cheaper / API cheaper / Not enough volume — with break-even reqs/day

Reference tab

6 sub-tables — fully transparent, every record source-linked where available:

Models (35 entries: Llama 4 Scout/Maverick, Qwen 3 family, DeepSeek V3/R1, Pixtral, BGE-M3, ...)
GPUs (30 entries: H100/H200/B200/B300, AMD MI300X/MI325X, Intel Gaudi 3, TPU v5p/Trillium, Groq LPU, Cerebras WSE-3, Apple M2/M3/M4 Ultra, ...)
Quantization (16 formats: FP16/BF16, GGUF Q2_K→Q8_0, GPTQ, AWQ, EXL2, FP8, NVFP4)
API pricing (13 models with input/cached/output + retired/current status)
Cloud GPU pricing (all GPUs with $/hr > 0 + typical providers — static curated)
🔴 Live pricing (real-time fetch from Azure + AWS + GCP + Vast.ai in parallel — LIVE badges + timestamps)

The long-context capacity planner (the differentiator)

This is the formula the Perplexity research brief called "the most important tokcalc should visibly expose":

$$ \text{KV bytes/request} = 2 \cdot L \cdot T \cdot H_{\text{kv}} \cdot D_h \cdot B $$

For dense attention, prefill cost grows superlinearly with context length:

$$ \text{prefill FLOPs} = \underbrace{2 \cdot N \cdot T}{\text{linear}} + \underbrace{T^2 \cdot H{\text{kv}} \cdot D_h \cdot L}_{\text{attention}} $$

tokcalc shows you:

Max concurrent users at 4K / 8K / 16K / 32K / 64K / 128K / 256K / 512K / 1M context
KV memory per request at each context length
Prefill time (with superlinear attention correction beyond 32K)
Required topology (Single GPU → TP×2/4/8 → TP×8 + Context Parallel)
RingAttention citation when CP is needed

Live GPU pricing (NEW)

The 🔴 Live pricing sub-tab in Reference fetches real-time GPU prices from 4 providers in parallel via Promise.allSettled:

Provider	API	Auth	Cache TTL
Azure	Retail Prices API	None (public)	24h
AWS	EC2 bulk pricing file	None (public)	24h
GCP	Cloud Billing Catalog API	`GCP_API_KEY` env var	24h
Vast.ai	Marketplace bundles API	None (public)	5m (spot prices change rapidly)

Each provider shows a LIVE badge with timestamp + cache state. The comparison table shows the cheapest price per GPU across all 4 providers (highlighted in emerald) + per-provider breakdown.

If any provider fails (e.g., GCP_API_KEY not set), the others still work — graceful degradation per provider.

The math, transparently

Every number above comes from a formula you can inspect. No black boxes.

Decode tokens/sec (memory-bandwidth bound)

$$ \text{decode tok/sec} \approx \frac{\text{HBM BW} \cdot \eta_{\text{mem}} \cdot \text{quant_eff}}{\text{model size}} $$

Where:

HBM BW = GPU memory bandwidth (e.g., 3350 GB/s for H100 SXM)
η_mem = 0.65 = typical real-world memory utilization (35% overhead)
quant_eff = dequantization efficiency multiplier (1.0 for FP16, 1.5 for FP8 on H100, 0.85 for INT4)
model size = active_params × bytes_per_param (uses ACTIVE params for MoE, not total)

Refs: PagedAttention paper (arxiv.org/abs/2309.06180)

Prefill tokens/sec (compute bound)

$$ \text{prefill tok/sec} \approx \frac{\text{GPU FLOPS} \cdot \eta_{\text{compute}}}{2 \cdot \text{active params}} $$

Where:

GPU FLOPS = dense FP16/BF16 TFLOPS (sparse values not used)
η_compute = 0.50 = typical compute utilization
Factor of 2 = one multiply + one add per parameter per token

For long context (>32K), the superlinear attention correction above applies.

Continuous batching multiplier (workload-specific)

$$ \text{aggregate tok/sec} = \text{decode tok/sec} \cdot \text{batch size} \cdot \text{continuous batching multiplier} $$

Critical caveat: There is no universal continuous batching multiplier. vLLM reported 14–24× vs HF Transformers (extreme), 2.2–2.5× vs TGI. SOSP paper finds 2–4× typical vs FasterTransformer/Orca. tokcalc defaults to a conservative 1.5× and lets you tune.

Refs:

Prompt caching economics (Anthropic / OpenAI)

For shared prefix of length $T_p$, suffix of length $T_u$, output $O$, cache hit rate $h$:

$$ \text{API input cost} = N \cdot \left[ (1-h) \cdot T_p \cdot P_{\text{write}} + h \cdot T_p \cdot P_{\text{read}} + T_u \cdot P_{\text{input}} \right] $$

Anthropic multipliers (verified 2025-2026):

5-minute cache write: 1.25× base input
1-hour cache write: 2.0× base input
Cache read: 0.1× base input (90% savings)

OpenAI: cached input discounted 50%, no separate write fee.

Refs:

Multi-GPU topology recommendation

$$ \text{total needed} = \text{model weights} + (\text{KV per request} \cdot \text{batch size}) $$

Walk the smallest topology that fits:

Single GPU: total ≤ VRAM × 1
Tensor Parallel ×2/4/8: total ≤ VRAM × N (weights + KV split evenly)
TP×8 + Context Parallel: total > VRAM × 8 — use RingAttention to shard KV across nodes

Refs: RingAttention paper

Self-host vs API break-even

$$ \text{self-host $/M tokens} = \frac{\text{GPU $/hr}}{3600 \cdot \text{effective tok/s} \cdot \text{utilization}} \cdot 10^6 $$

$$ \text{break-even req/day} = \frac{\text{monthly self-host cost}}{30 \cdot \text{API cost per request}} $$

The decisive term is effective utilization — not peak throughput. A GPU running at 10% utilization pays 10× more per token than the theoretical minimum.

Comparison with adjacent tools

Capability	induwara / techfuelhq / pcmasterstudio	gigagpu / kickllm / cloudparity	HF Open LLM Leaderboard / MLPerf	tokcalc
Model × GPU × quant tok/s	✓	some	some	✓
Model-fit / VRAM	basic	rare	rare	✓
KV-cache by context + concurrency	—	—	implicit	✓
Prefill vs decode split (TTFT/ITL)	—	—	engine-specific	✓
Continuous batching / paged attention	—	—	docs only	✓
Long-context (128K–1M) planning	—	—	—	✓
Topology recommendation (TP/CP)	—	—	partial	✓
Prompt-cache economics	—	partial API only	—	✓
Reasoning tokens (o1/R1/Claude thinking)	—	—	—	✓
API vs self-host break-even	some	✓	—	✓
Live cloud GPU pricing (4 providers)	—	—	—	✓
Observed benchmark calibration	—	—	—	✓
MCP server for AI agents	—	—	—	✓
Public hosted MCP endpoint + self-serve keys	—	—	—	✓
Transparent formulas / open source	mixed	usually no	mixed	✓
Cited benchmark evidence per config	rare	rare	✓ (not planning)	✓ (in progress)
Shareable URL per config	—	—	—	✓
Copy result as Markdown	—	—	—	✓

Roadmap

Shipped

✅ 35 models, 30 GPUs, 16 quantization formats
✅ Continuous batching, reasoning tokens, prompt caching
✅ TTFT/ITL split, long-context superlinear attention
✅ Long-context capacity planner + topology recommendation
✅ Build-vs-Buy calculator (13 API providers with retired/current status)
✅ Reference catalog (6 sub-tables including live pricing)
✅ Live GPU pricing (Azure + AWS + GCP + Vast.ai parallel fetch with LIVE badges)
✅ Observed benchmark calibration (paste vLLM/SGLang/TRT-LLM JSON → verdict)
✅ MCP server v0.2.0 — 7 read-only tools, stdio + Streamable HTTP transports
✅ Public hosted MCP endpoint at tokcalc.vercel.app/api/mcp with bearer auth + Upstash Redis rate limiting
✅ Self-serve API key generation at tokcalc.vercel.app/mcp (email → instant key)
✅ 4 SEO landing pages: /compare/h100-vs-h200, /compare/gguf-q4-k-m-vs-q5-k-m, /self-host-vs-openai-api, /mcp (install docs)
✅ Share URL + localStorage persistence
✅ Copy result as Markdown (for GitHub issues / Slack / docs)
✅ Dark mode toggle
✅ Plain-English glossary (28 terms with hover tooltips)
✅ OG image + Twitter card + social metadata
✅ Plausible Analytics (privacy-friendly) + Sentry error monitoring

Next 30 days

⏳ GitHub Action (tokcalc/plan PR comment)
⏳ i18n: Chinese, Japanese, Korean
⏳ MCP server v0.2.1: Fix empty inputSchema in tools/list response (zod-to-json-schema serialization issue)
⏳ MCP server v0.3.0: OAuth 2.1 with PKCE for multi-user auth

Next 90 days

⏳ Workload-trace / SLO capacity planner (prompt/output/arrival distributions, p50/p95 TTFT/ITL)
⏳ P/D disaggregation planner (separate prefill + decode pools)
⏳ Cache-aware economics (prefix-sharing distribution, multi-turn/agent traces)
⏳ Engine-aware presets (vLLM / SGLang / TensorRT-LLM / llama.cpp) — UI integration
⏳ Versioned price + benchmark provenance system

Long-term

🔮 Agentic workflow calculator (multi-turn + tool calls + growing context)
🔮 Multi-LoRA capacity planner (Punica / S-LoRA economics)
🔮 VLM image-token accounting (per-model patch/tile tokenization)
🔮 Embedding model mode (vectors/sec, separate workload)
🔮 Training/fine-tuning estimator (LoRA / QLoRA / full-SFT FLOPs)
🔮 Energy / carbon per million tokens (region-specific grid intensity)

Open core model

tokcalc is open core — the calculator and catalog are open source; the cloud/data/team features are paid.

Asset	License	Notes
Source code	Apache 2.0	This repo. Free to use, modify, distribute
Model/GPU/quant catalog	CC0 1.0	Public domain data. Anyone can use, no attribution required
Benchmark provenance data	CC-BY-SA 4.0	Anyone can use, but must attribute + share-alike
Documentation	CC-BY 4.0	Attribution required if copied
"tokcalc" name + logo	Trademark	Even without formal registration, common-law rights apply
Cloud SaaS layer	Proprietary	Real-time pricing API, benchmark DB, team workspaces (coming soon)

Why this structure

Trust: Open-source formulas build credibility vs opaque competitors
Community: Contributors can submit models, GPUs, quants, benchmarks
Defensibility: Trademark + cloud features + URL-share viral loop protect against forks
Revenue: Cloud tier funds ongoing development + pricing/benchmark data maintenance

Contributing

We welcome contributions! See CONTRIBUTING.md for:

How to add a model (with HuggingFace config.json as source)
How to add a GPU (with critical guardrails for B200/B300 null FP16 fields)
How to add a quantization format (with measured file sizes as source)
How to submit a benchmark (3-tier confidence model)
How to improve a formula (cite the source, no magic numbers)

Most-needed contributions

🟢 New models (Qwen 3, Mistral Large 4, Llama 4 variants as they release)
🟢 New GPUs (B300, AMD MI400, Apple M5 Ultra when shipping)
🟢 New quantization formats (mxFP8, MXFP4, BitNet 2)
🟢 Real benchmark data (run vLLM benchmarks and submit with provenance)
🟢 Translations (especially Chinese, Japanese, Korean)

MCP server — use tokcalc from AI agents

tokcalc ships an MCP (Model Context Protocol) server that lets AI agents (Cursor, Claude Desktop, Cline) call tokcalc during design reviews.

npm package: @tokcalc/mcp-server (latest: v0.2.0) Public endpoint: https://tokcalc.vercel.app/api/mcp (Streamable HTTP, bearer auth, rate-limited) Install docs: https://tokcalc.vercel.app/mcp

7 read-only tools

Tool	What it does
`estimate_capacity`	VRAM/KV/throughput/latency/cost for one config
`compare_gpus`	Ranked GPU comparison for one workload
`recommend_topology`	TP/CP topology recommendation
`estimate_api_vs_self_host`	Break-even analysis
`list_models`	Discover supported model IDs
`list_gpus`	Discover supported GPU IDs
`get_mlperf_benchmarks`	Curated MLPerf Inference v4.1 audited reference configs

All tools are read-only — no side effects, no cloud credentials, no deployments.

Option A: Public hosted endpoint (recommended — no install)

Point Cursor or Claude Desktop at the public endpoint. Get an instant API key at /mcp:

config.json

{
  "mcpServers": {
    "tokcalc": {
      "url": "https://tokcalc.vercel.app/api/mcp",
      "headers": { "Authorization": "Bearer <your-api-key>" }
    }
  }
}

Features:

✅ Bearer API key auth (constant-time comparison via crypto.timingSafeEqual)
✅ KV-backed rate limiting via Upstash Redis (30 req/min per IP, 120 req/min per key)
✅ Self-serve key generation — enter email at /mcp, get instant key (5 per IP per day limit)
✅ Keys expire after 90 days of inactivity
✅ SHA-256 key hashing (never stored raw)

Option B: Self-hosted stdio (for local Cursor / Claude Desktop)

config.json

{
  "mcpServers": {
    "tokcalc": {
      "command": "npx",
      "args": ["-y", "@tokcalc/mcp-server"]
    }
  }
}

No API key needed for stdio (local install). Cursor spawns the process via npx.

Option C: Self-hosted HTTP (for air-gapped / custom infra)

Terminal

npm install -g @tokcalc/mcp-server
tokcalc-mcp-http  # defaults to port 3000

Set MCP_API_KEY, UPSTASH_REDIS_REST_URL, UPSTASH_REDIS_REST_TOKEN env vars for auth + rate limiting.

Example agent prompt

"I need to serve Llama 3.3 70B at 32K context for 50 concurrent users. What GPU topology do you recommend, and how much will it cost per month?"

The agent calls list_models → list_gpus → recommend_topology → estimate_capacity → returns a structured plan with VRAM, throughput, latency, cost, and confidence.

Tech stack

Framework: Next.js 16 with App Router (Turbopack)
Language: TypeScript 5
Styling: Tailwind CSS 4 + shadcn/ui (New York)
Charts: Recharts
State: React hooks (useState + useEffect + useMemo)
Theme: next-themes (dark mode default)
MCP SDK: @modelcontextprotocol/sdk v1.30.1 (Streamable HTTP transport, stateless mode)
Auth: crypto.timingSafeEqual (constant-time bearer key validation)
Rate limiting: @upstash/ratelimit + @upstash/redis (sliding window, 30/min per IP, 120/min per key)
Analytics: Plausible (privacy-friendly, no cookies)
Error monitoring: Sentry
Database: None for core; Upstash Redis for API key storage + rate limiting

Project structure

Code

tokcalc/
├── src/
│   ├── app/
│   │   ├── api/
│   │   │   ├── mcp/route.ts          # Public MCP endpoint (Streamable HTTP)
│   │   │   ├── keys/route.ts         # Self-serve API key generation
│   │   │   └── pricing/              # Live GPU pricing proxies
│   │   │       ├── aws/route.ts
│   │   │       ├── azure/route.ts
│   │   │       ├── gcp/route.ts
│   │   │       └── vast-ai/route.ts
│   │   ├── compare/                  # SEO landing pages
│   │   │   ├── h100-vs-h200/page.tsx
│   │   │   └── gguf-q4-k-m-vs-q5-k-m/page.tsx
│   │   ├── self-host-vs-openai-api/page.tsx
│   │   ├── mcp/page.tsx              # MCP install docs + self-serve key form
│   │   ├── page.tsx                  # Main calculator
│   │   └── layout.tsx
│   ├── lib/
│   │   ├── mcp/
│   │   │   ├── core.ts              # UNIFIED tool core — all 11 tools, both surfaces
│   │   │   ├── server.ts            # thin shim → core (createMcpServer factory)
│   │   │   ├── mcp-version.ts       # MCP_SERVER_VERSION — single source of truth
│   │   │   ├── mcp-store.ts         # per-install store (spec cache + calibration)
│   │   │   └── auth.ts               # validateApiKey (Redis + env var)
│   │   ├── token-calc.ts             # Core calculation engine
│   │   ├── mlperf-curated.ts         # 9 MLPerf v4.1 reference configs
│   │   ├── price-schema.ts           # PriceRecord type + helpers
│   │   ├── benchmark-schema.ts       # BenchmarkRecord + CalibrationResult
│   │   ├── engine-presets.ts         # 5 engine presets (vLLM/SGLang/TRT-LLM/llama.cpp/generic)
│   │   └── url-state.ts              # URL hash + localStorage persistence
│   └── components/
│       ├── benchmark-import.tsx      # Paste JSON → calibrate against estimate
│       └── confidence-badge.tsx      # 🟢🟡⚪ confidence dots
├── mini-services/
│   └── mcp-server/                   # Standalone npm package (for stdio + self-hosted HTTP)
│       ├── server.ts                 # thin shim → src/lib/mcp/core.ts (no duplicated handlers)
│       ├── auth.ts                   # Bearer auth middleware (Node http)
│       ├── rate-limit.ts             # Upstash rate limiter (Node http)
│       ├── http.ts                   # Standalone HTTP entry (tokcalc-mcp-http)
│       ├── index.ts                  # Stdio entry (tokcalc-mcp-server)
│       └── package.json              # @tokcalc/mcp-server npm package
├── public/
│   ├── sitemap.xml                   # 5 URLs (/, /mcp, 3 compare pages)
│   ├── og.png                        # OG image
│   └── manifest.json                 # PWA manifest
├── server.json                       # MCP registry entry (packages + remotes)
├── package.json                      # Web app deps
└── next.config.ts

Acknowledgments

tokcalc builds on the work of:

vLLM team — PagedAttention, continuous batching (arxiv.org/abs/2309.06180)
llama.cpp / ggml-org — GGUF format and quantization variants
MLPerf / MLCommons — standardized inference benchmark methodology
Stanford HELM — efficiency-aware model evaluation framework
Anthropic / OpenAI / Google — published prompt-caching pricing rules
Model Context Protocol team — MCP Streamable HTTP transport spec (2025-03-26)
NVIDIA / AMD / Intel / Google / Groq / Cerebras — published hardware specs
Upstash — Redis + rate limiting infrastructure

Every formula has a citation. Every model/GPU/quant entry has a source URL where available. If you spot an unsourced claim, please open an issue.

Star history

Live demo · MCP install docs · Public MCP endpoint · npm package · Contributing · License · Code of Conduct

Made with care by the tokcalc community. Apache 2.0 licensed.