← Home

Cost across cloud + local

Paste anything — a prompt, source file, transcript, email draft. We count tokens, then show you the cost across 11 cloud providers and your local rig, side by side. Same data for the same input. Decide where each workload belongs.

11 providers across 3 tiers. Prices verified May 2026. Token counts are content-aware approximations (±15%) — provider price differences dwarf tokenizer accuracy, so this is enough to make a decision. URL updates as you change — share by copy.

§ Editorial stance
Cloud APIs from Anthropic, OpenAI, Google, DeepSeek and the open-weight hosters are world-class — modern local AI exists downstream of their research. They're the right call for frontier-quality work, day-one access to new models, and workloads where you don't want to own hardware. Local is the right call for high-volume routine work, privacy-sensitive content, and offline capability. This page is a calculator, not advocacy — it shows you where each option lands for your specific workload so you can pick deliberately, not by default.

Calculator

Start with the sample text or paste your own. Pick a local rig + model from your catalog and the cloud providers you want to compare against. Everything is reactive.

§ Paste a prompt, code, or transcript
457 chars·detected: Code (TS / JS / Python)·~3.2 chars/token
Estimated tokens: 143 in · 72 out
§ Compare against (optional)
§ Cloud providers to compare
4 selected · prices verified 2026-05
Frontier
Mid-tier
Open-weight (hosted)
At 1,500 runs per month
$3.23on GPT-5 (most expensive)
Cheapest cloud: $0.177 on DeepSeek V3 (API).
§ Cost spectrum — per month
longer bar = more expensive · color = provider tier
GPT-5$3.2GPT-5: $3.2 per month · $0.0022 per runClaude Sonnet 4.5$2.3Claude Sonnet 4.5: $2.3 per month · $0.0015 per runLlama 3.3 70B (Together)$0.28Llama 3.3 70B (Together): $0.28 per month · $0.0002 per runDeepSeek V3 (API)$0.18DeepSeek V3 (API): $0.18 per month · $0.0001 per runCost over 1,500 runs (per month)
Frontier (premium)
Mid-tier
Open-weight (hosted)
§ Receipt — 1,500 runs at 143 in / 72 out tokens each
ProviderModel$/M in$/M outPer runper monthvs cheapest
DeepSeek
DeepSeek V3 (API)
Mid-tier
$0.27$1.10$0.00012$0.177
cheapest
Together AI
Llama 3.3 70B (Together)
Open-weight
$0.88$0.88$0.00019$0.2841.6×
Anthropic
Claude Sonnet 4.5
Frontier
$3.00$15.00$0.00151$2.2612.8×
OpenAI
GPT-5
Frontier
$5.00$20.00$0.00216$3.2318.3×
Range: $0.177 → $3.23 (18.3× spread). Prices verified May 2026. Token counts approximate ±15%.

Where to go from here

Stack Builder →

Once you know it's worth going local: pick the whole rig + runtime + model + install script in one wizard.

Quant Advisor →

The local cost above assumes Q4_K_M. Drill into which quant fits your specific hardware × model × context.

TCO calculator →

The break-even number above is volume-driven. Switch view to tune utilization / electricity / amortization assumptions.

Stream Visualizer →

Cost is one side of the trade — speed is the other. Watch tokens stream at the rate your rig would produce them, side-by-side with the cloud provider.

Claude Code → local →

Step-by-step guide to swap Claude Sonnet for a local model in your Claude Code workflow. Real commands.