Guides → Playground & Guide → Training Run Cost - GPU Cost + Wall-Clock for a Training or Fine-Tune Run

Training Run Cost - GPU Cost + Wall-Clock for a Training or Fine-Tune Run

🚀 Open the calculator

Meet Dana Okafor. ML lead pricing a fine-tune before asking for budget. "We want to fine-tune a 7B model on ~10B tokens. What does the GPU run actually cost, and how long does it take on one node vs eight?"

🔥 Finance wants a number before approving the cluster. 'A few thousand' vs 'six figures' depends entirely on model size and token count - and nobody on the team can ballpark it.

The story

Training cost is compute, divided by your GPUs, priced by the hour. The compute is roughly 6 · params · tokens · epochs FLOPs. Your GPUs chew through it at their peak FLOPS times a real-world utilization (MFU, ~40%).

Adding nodes makes the run finish faster but doesn't make it cheaper - the GPU-hours (and the bill) scale with the work, not the wall-clock. Eight nodes finishes in an eighth the time at roughly the same dollars.

Dana's 7B-on-10B-tokens fine-tune on one 8×H100 node: a few thousand dollars and a day-and-a-half of wall-clock. The same engine prices a 70B continued-pretrain on a cluster into six figures - so she can right-size before committing.

🎮 Playground

Training Run Cost Playground

Here are the inputs that move the result the most. Play with the sliders and check it out. The number updates live.

What will a training run cost?

A one-time training run: params × tokens on a GPU cluster. GPU type / commitment / epochs default (8×H100 p5, on-demand, 1 epoch).

Estimated run cost
—

💡FLOPs ≈ 6 × params × tokens; more instances finish faster but the run cost tracks total GPU-hours. This is a per-run cost.

Ready to run the numbers?

Open the full calculator. Pick a model, enter your tokens, see per-call, daily, monthly, and annual cost.

🚀 Open the full calculator →

GPU instances by per-GPU-hour price

Verified 13 hours ago
  1. 1
    GPT-5 Mini
    $0.250 in · $2.00 out ·
  2. 2
    gpt-5.1-codex-mini
    $0.250 in · $2.00 out ·
  3. 3
    Command
    $1.00 in · $2.00 out ·

About this calculator: Training Run Cost - GPU Cost + Wall-Clock for a Training or Fine-Tune Run

Estimate what one training or fine-tune run costs - by model size, training tokens, GPU type (H100/A100/A10G), node count, and commitment (on-demand/reserved/spot).

🎛 Inputs you control

Each input shapes the cost. Click an input on the calculator to set it. The explanations below match the live calculator field by field.

▸ Model size (B params): Parameter count of the model being trained, in billions.
How to choose: Use the target model size; compute scales with this.
▸ Dataset tokens (B): Total tokens the run trains on, in billions (includes multiple epochs).
How to choose: Tokens times epochs; more data or passes raises cost proportionally.
▸ Training instances: Number of GPU instances / accelerators running the job in parallel.
How to choose: More instances finish sooner at similar total compute — set for your deadline.
▸ GPU instance: The GPU instance type the run executes on.
How to choose: Match your accelerators; it sets the hourly rate behind the run cost.
▸ Commitment: Pricing commitment: on-demand, reserved, or spot.
How to choose: Spot is cheapest but can interrupt; reserved discounts steady long runs.
▸ Epochs: How many full passes over the dataset.
How to choose: More epochs multiply tokens processed and cost; fine-tunes often need only 1 to 3.
▸ MFU (0 to 1): Model FLOPs utilization — how efficiently the GPUs are used.
How to choose: Real runs land ~0.3 to 0.5; higher MFU means fewer GPU-hours for the same work.
📋 Typical values & starting points Don’t know a value yet? Start with these broad, sourced ballparks — the calculator’s ▾ Typical menus pre-load the same options.

Context: GPU market Jul 2026: H100 neocloud ~$2.3-3.1/hr on-demand (hyperscaler ~$7.4-8), spot from ~$0.34-1.5; B200 ~$5.5-6 on-demand, spot ~$2.12, 36-mo reserved ~$2.25. Typical fine-tune MFU 30-50%.

Input Default Typical ballparks
paramsB moves the needle 7 Small · 7B = 7 · Mid · 13B = 13 · Large · 70B = 70
datasetTokensB moves the needle 10 Fine-tune corpus · 0.05B (50M tok) = 0.05 · Domain corpus · 1B = 1 · Continued pretrain · 10B = 10 · Chinchilla-style · 20× params = 140
epochs 1 Single pass · 1 = 1 · Typical SFT · 3 = 3 · Small-set SFT · 5 = 5
mfu 0.4 Unoptimized · 0.3 = 0.3 · Typical · 0.4 = 0.4 · Well-tuned · 0.5 = 0.5
instanceCount moves the needle 1 Single node · 1 = 1 · Small cluster · 4 = 4 · 8-GPU node · 8 = 8
gpuKey aws:p5.48xlarge —
commitment onDemand —
managed false —

Ballparks are broad industry starting points (sourced ranges; * = rough estimate) — your result gets more accurate as you replace them with measured numbers. Try them live in the calculator; API & agent users get the same data from the MCP resource aicost://input-reference/training-run-cost.

Most popular

Everything above is the 80% case. The last 20% is where the money is.

The gaps we just listed are real, and they are the expensive ones: your actual prompts, your switching cost, your MLOps overhead. An AICost expert spends the hour on your AI and cloud costs, not a generic playbook. You leave with a written report: the way forward, in 30 days of concrete steps.

  • An hour with the people who built the engines
  • Report the same day
  • The fee credits toward any AICost plan
  • Two slots a week
Book a Solution Session: $299 → or $99 for small business →

Not sure yet? The $39 AICost Blueprint credits toward a Session, and the Session fee credits toward any plan. You never pay twice for the same ground. See all pricing →

Ready to run your own numbers?

You have seen the shape of it. Open the calculator with your model, your tokens, your volume.

🚀 Open the full calculator →

Methodology

Editorial gate
8-layer defense, see aicost.ai/ai-cost-economics
Last verified
7/20/2026, 8:00:00 PM

Author: Subu Vdaygiri, Founder & CEO of CloudIntelligence.ai. 17 years Fortune 100 (Ingram Micro, Siemens). Wharton CTO program · Kellogg CPO program · 10× AWS+Azure certified.

3 years of pricing history

Why this matters: pricing for major vendors has dropped 40-90% in the last 24 months. A budget set 12 months ago is probably wrong by 30%+.

View 3-year history for →
📖 Data sources & methodology 158 text models · 9 embeddings · 37 vision · 55 audio · 8 vector DBs across 10 vendor pages · last verified 2026-07-21

Methodology

  • All prices are USD per 1 million tokens, current as of 2026-07-21.
  • Vendor-published values have no mark. Inferred/extrapolated values are marked with * and listed below.
  • Batch API discounts are 50% off standard rates across providers that offer Batch mode.
  • Prompt caching discounts vary by provider (typically 80-90% off cached input tokens).
  • Regional data-residency surcharges (Anthropic 1.1x, OpenAI 1.1x, Google regional tiers) are NOT included in base rates.
  • Long-context pricing tiers apply when input exceeds model threshold.
  • Embedding prices are input-only (no output tokens generated).

Primary sources

Last-verified date is the most recent successful daily snapshot (aicost_pricing_snapshots) or, when no snapshot exists yet, the latest successful crawler run (aicost_crawler_runs). 10 of 10 vendors are currently verified. Aggregator services (TokenCost, AI Pricing Guru, etc.) are not listed.

Anthropic
2026-07-21
https://www.anthropic.com/pricing
Daily snapshot since Sep 2023 · 624 days captured
Anthropic Docs
2026-07-21
https://platform.claude.com/docs/en/about-claude/pricing
Daily snapshot since Sep 2023 · 624 days captured
OpenAI
2026-07-21
https://openai.com/api/pricing/
Daily snapshot since Sep 2023 · 625 days captured
Google AI
2026-07-21
https://ai.google.dev/gemini-api/docs/pricing
Daily snapshot since Dec 2023 · 600 days captured
Google Vertex
2026-07-21
https://cloud.google.com/vertex-ai/generative-ai/pricing
Daily snapshot since Dec 2023 · 600 days captured
DeepSeek
2026-07-21
https://api-docs.deepseek.com/quick_start/pricing
Daily snapshot since May 2024 · 539 days captured
xAI
2026-07-21
https://x.ai/api
Daily snapshot since Nov 2024 · 457 days captured
Mistral
2026-07-21
https://mistral.ai/pricing
Daily snapshot since Dec 2023 · 598 days captured
Cohere
2026-07-21
https://cohere.com/pricing
Daily snapshot since Sep 2023 · 624 days captured
Voyage AI
2026-07-21
https://docs.voyageai.com/docs/pricing

Inferred values (marked with * in calculator tables)

Derived from industry conventions, not directly published by the vendor. Typical conventions: cached input = 10% of base (90% off), Batch API = 50% of base (50% off).

Vendor / Model Field Why it’s inferred
Anthropic — Claude Sonnet 4.6 cachedInput Derived at 10% of input rate — Anthropic publishes 90% cache-hit discount on this tier.
Anthropic — Claude Sonnet 4.5 cachedInput Derived at 10% of input rate; same 90% cache-hit convention as Sonnet 4.6.
Anthropic — Claude Sonnet 4.5 batchInput Derived at 50% of standard input — Anthropic documents uniform 50% Batch discount.
Anthropic — Claude Sonnet 4.5 batchOutput Derived at 50% of standard output — Anthropic documents uniform 50% Batch discount.
Anthropic — Claude Haiku 4.5 cachedInput Derived at 10% of input rate — Anthropic 90% cache-hit discount convention.
OpenAI — GPT-5.4 Mini cachedInput Derived at 10% of input — OpenAI documents automatic 90% discount on cache hits across GPT-5.x tier.
OpenAI — GPT-5.4 Nano cachedInput Derived at 10% of input — OpenAI 90% cache-hit convention.
OpenAI — GPT-5.4 Nano batchInput Derived at 50% of input — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.4 Nano batchOutput Derived at 50% of output — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.4 Pro cachedInput Derived at 10% of input — OpenAI 90% cache-hit convention.
OpenAI — GPT-5.4 Pro batchInput Derived at 50% of input — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.4 Pro batchOutput Derived at 50% of output — OpenAI Batch API uniform 50% discount.
OpenAI — GPT-5.2 cachedInput Derived at 10% of input; no residency uplift.
OpenAI — GPT-5.2 batchInput Derived at 50% of input.
OpenAI — GPT-5.2 batchOutput Derived at 50% of output.
OpenAI — GPT-5 cachedInput Derived at 10% of input.
OpenAI — GPT-5 batchInput Derived at 50% of input.
OpenAI — GPT-5 batchOutput Derived at 50% of output.
OpenAI — GPT-5.5 Pro cachedInput Derived at 10% of input — OpenAI does not publish a cached rate for *-pro models; using the family convention.
OpenAI — GPT-5.5 Pro batchInput Derived at 50% of input.
OpenAI — GPT-5.5 Pro batchOutput Derived at 50% of output.
OpenAI — GPT-5.2 Pro cachedInput Derived at 10% of input — pro-tier convention.
OpenAI — GPT-5.2 Pro batchInput Derived at 50% of input.
OpenAI — GPT-5.2 Pro batchOutput Derived at 50% of output.
OpenAI — GPT-5.1 batchInput Derived at 50% of input.
OpenAI — GPT-5.1 batchOutput Derived at 50% of output.
OpenAI — GPT-5 Pro batchInput Derived at 50% of input.
OpenAI — GPT-5 Pro batchOutput Derived at 50% of output.
OpenAI — GPT-5 Nano cachedInput Derived at 10% of input.
OpenAI — GPT-5 Nano batchInput Derived at 50% of input.
OpenAI — GPT-5 Nano batchOutput Derived at 50% of output.
Google — Gemini 3 Flash cachedInput Derived at 10% of input — Google caching discount convention ~90%.
Google — Gemini 3.1 Flash-Lite cachedInput Derived at 10% of input — Google caching convention.
Google — Gemini 3.1 Flash-Lite batchInput Derived at 50% of input — Google Batch API uniform 50% discount.
Google — Gemini 3.1 Flash-Lite batchOutput Derived at 50% of output — Google Batch API uniform 50% discount.
Google — Gemini 2.5 Pro cachedInput Derived at 10% of input.
Google — Gemini 2.5 Flash cachedInput Derived at 10% of input.
Google — Gemini 2.5 Flash-Lite cachedInput Derived at 10% of input — Google caching convention.
Google — Gemini 2.5 Flash-Lite batchInput Derived at 50% of input — Google Batch API uniform 50% discount.
Google — Gemini 2.5 Flash-Lite batchOutput Derived at 50% of output — Google Batch API uniform 50% discount.
Google — Gemini 2.0 Flash cachedInput Derived at 25% of input per Google 2.0 family caching rates.
Google — Gemini 2.0 Flash batchInput Derived at 50% of input — Google Batch API uniform 50% discount.
Google — Gemini 2.0 Flash batchOutput Derived at 50% of output — Google Batch API uniform 50% discount.
Google — Gemini 2.0 Flash-Lite cachedInput Derived at 10% of input — Google caching convention.
Google — Gemini 2.0 Flash-Lite batchInput Derived at 50% of input — Google Batch API uniform 50% discount.
Google — Gemini 2.0 Flash-Lite batchOutput Derived at 50% of output — Google Batch API uniform 50% discount.
xAI — Grok 4 (legacy) cachedInput Extrapolated at 25% of base.

Pricing is cross-verified against the LiteLLM community registry when available. Daily snapshots are kept in aicost_pricing_snapshots; every change is logged to aicost_price_changelog with old & new values for full audit trail. Read the full methodology →