← Back to counter

What is the cheapest AI model?

Updated 2026-05-31 · By Clinton Patrick · Methodology

The short answer

As of April 27, 2026, GPT-5 Nano at $0.05 per million input tokens / $0.40 per million output tokens is the cheapest exact-tokenizer model from a major provider.

For multimodal workloads or longer context, Gemini 2.5 Flash-Lite ($0.10 / $0.40) is the closest GA alternative, same output price, double the input rate, but with a 1M-token context window vs GPT-5 Nano's 400K.

Cheapest models, ranked

Sorted by per-1M-call cost on a typical 1,000-token input + 200-token output workload:

"Cheap" depends on what you need

The cheapest model isn't always the right answer. Consider:

• Tokenizer accuracy. OpenAI, Anthropic, and Google models in this list have exact counts. Open-weights models (Llama, Mistral, DeepSeek, Qwen) are estimated to ~±3%.
• Capability gap. GPT-5 Nano and Gemini 2.5 Flash-Lite are designed for high-volume routing, classification, and extraction. They will fall behind on hard reasoning. For mid-tier tasks, GPT-5 Mini or Claude Haiku is the right call.
• Output-heavy vs input-heavy. Llama 8B charges the same per-token rate for input and output, which makes it interesting for output-heavy generation but worse for input-heavy RAG. GPT-5 Nano's 8× output-to-input ratio rewards short replies.
• Context length needs. GPT-5 Nano caps at 400K tokens. Gemini 2.5 Flash-Lite handles 1M. Llama 4 Scout (when added) handles 10M. Choose based on your typical input length.
• Preview vs GA. Gemini 3 family and OpenAI's "Pro" tiers are still in Preview/early-access, pricing or behavior could shift. GA models give pricing stability.

Cheaper than this list

If you really want sub-$0.05 per million tokens:

• Self-host Llama 3.1 8B or Qwen 2.5 7B on your own GPU. Hardware cost only, no per-token bill.
• Run prompt caching on any of the above, cached input drops to ~10% of standard rate (so GPT-5 Nano cached input is $0.005/M).
• Use the Batch API on OpenAI / Anthropic, flat 50% discount for asynchronous workloads.

Get a real cost comparison

Paste your prompt into the counter, it shows the actual token count and per-call cost across every model, so you can choose by total cost on your workload instead of by per-million headline.

Try this on every model

• Claude Opus 4.8 $5.00/$25.00
• Claude Opus 4.8 (Fast Mode) $10.00/$50.00
• Claude Sonnet 4.6 $3.00/$15.00
• Claude Haiku 4.5 $1.00/$5.00
• GPT-5.5 $5.00/$30.00
• GPT-5.5 Pro $30.00/$180.00
• GPT-5.4 $2.50/$15.00
• GPT-5.4 Mini $0.75/$4.50
• GPT-5.4 Nano $0.20/$1.25
• GPT-5.4 Pro $30.00/$180.00
• GPT-5.3 $1.75/$14.00
• GPT-5.2 $1.75/$14.00
• GPT-5.2 Pro $21.00/$168.00
• GPT-5.1 $1.25/$10.00
• GPT-5 $1.25/$10.00
• GPT-5 Mini $0.25/$2.00
• GPT-5 Nano $0.05/$0.40
• GPT-5 Pro $15.00/$120.00
• GPT-4.1 $2.00/$8.00
• GPT-4.1 Mini $0.40/$1.60
• GPT-4.1 Nano $0.10/$0.40
• o3 $2.00/$8.00
• o3-mini $1.10/$4.40
• o3-pro $20.00/$80.00
• o4-mini $1.10/$4.40
• GPT-4o $2.50/$10.00
• GPT-4o mini $0.15/$0.60
• GPT-4 Turbo $10.00/$30.00
• Gemini 3.1 Pro $2.00/$12.00
• Gemini 3 Flash $0.50/$3.00
• Gemini 3.1 Flash-Lite $0.25/$1.50
• Gemini 2.5 Pro $1.25/$10.00
• Gemini 2.5 Flash $0.30/$2.50
• Gemini 2.5 Flash-Lite $0.10/$0.40
• Llama 3.3 70B $0.88/$0.88
• Llama 3.1 405B $3.50/$3.50
• Llama 3.1 70B $0.59/$0.79
• Llama 3.1 8B $0.18/$0.18
• Mistral Large $2.00/$6.00
• DeepSeek V3 $0.27/$1.10
• DeepSeek V3.1 $0.60/$1.70
• DeepSeek R1 $3.00/$7.00
• Qwen 2.5 72B $0.90/$0.90
• Qwen 2.5 Coder 32B $0.80/$0.80
• Qwen3 Coder 480B $2.00/$2.00
• GLM-5.1 $1.40/$4.40

Try the live counter →