Calculate Your Exact Costs
Enter your token usage and see exactly how much each of these 15 models costs for your workload.
Open Cost Calculator →Price Tier Breakdown
These 15 models split into three clear tiers based on input pricing. Here's what you get at each level.
Under $0.15/M input
- GPT-oss 20B at $0.08/$0.35 is the cheapest model
- GPT-oss 20B — $0.08/$0.35 · 128K context · OpenAI's open-source entry
- Llama 3.1 8B — $0.10/$0.10 · 128K context · Open source, lowest total cost
- Gemini 2.5 Flash-Lite — $0.10/$0.40 · 1M context · Google's balanced budget option
- DeepSeek V4 Flash — $0.14/$0.28 · 1M context · Best budget coding model
Best for: high-volume classification, simple Q&A, data extraction, embedding pipelines
$0.15 — $0.30/M input
- GPT-oss 120B — $0.15/$0.60 · 128K context · Strong general-purpose
- GPT-4o mini — $0.15/$0.60 · 128K context · OpenAI's budget workhorse
- Mistral Small 4 ($0.15/$0.60) · 128K context · EU data sovereignty
- Llama 4 Scout — $0.18/$0.59 · 1M context · Open source, MIT license
- DeepSeek V3.2 — $0.23/$0.34 · 128K context · Proven production model
- Llama 4 Maverick — $0.27/$0.85 · 1M context · Open source flagship
Best for: chatbots, content generation, RAG pipelines, code assistance
$0.40 — $1.00/M input
- DeepSeek V4 Pro — $0.435/$0.87 · 1M context · Best value for complex tasks
- Mistral Large 3 — $0.50/$1.50 · 262K context · Strong at RAG and retrieval
- Command R — $0.50/$1.50 · 128K context · Cohere's enterprise RAG model
- Kimi K2.6 — $0.95/$4.00 · 256K context · Excellent reasoning capabilities
Best for: code generation, complex analysis, RAG, nuanced writing
Use Case Recommendations
Different tasks need different models. Here's the best sub-$1 model for each major use case.
Chatbot
$0.14/$0.28 — cheapest model that handles multi-turn conversations naturally. Used in production by thousands of apps.
Code Generation
$0.435/$0.87 — outperforms GPT-4o on coding benchmarks at 80% less cost. Best value coding model under $1.
RAG Pipeline
$0.50/$1.50 — purpose-built for retrieval-augmented generation with strong context following and factual accuracy.
Content Writing
$0.95/$4.00 — excellent reasoning and long-form generation. The most capable writing model under $1/M input.
Data Extraction
$0.10/$0.10 — lowest total cost at $0.20/M. Perfect for high-volume structured extraction tasks.
Multilingual
$0.15/$0.60 — strong multilingual support with EU data sovereignty. Handles 30+ languages well.
No Vendor Lock-in
$0.18/$0.59 — open source MIT license, self-hostable, 1M context window. Full control over your stack.
Highest Volume
$0.075/$0.30 — cheapest input price of any model. When you need to process millions of tokens daily.
Track Every Dollar with APIpulse
Set cost alerts, compare models in real-time, and optimize your API spend across all 15 budget models. Free forever.
Try APIpulse Free →Provider Breakdown
Eight providers offer models under $1/M input in July 2026. Here's how they compare.
- Google — 2 models. Cheapest input price with Gemini Flash Lite ($0.075). Both models have 1M context windows.
- OpenAI — 3 models. GPT-oss 20B and 120B are open-source. GPT-4o mini is the industry standard budget model.
- Meta — 3 models. Llama 3.1 8B is the lowest total cost. Llama 4 Scout and Maverick both have 1M context, MIT license.
- DeepSeek — 3 models. V4 Flash ($0.14) is best for coding. V3.2 is proven in production. V4 Pro is best value for complex tasks.
- Mistral — 2 models. EU-based with data sovereignty. Mistral Small 4 competes directly with GPT-4o mini.
- Cohere — 1 model. Command R at $0.50 is purpose-built for RAG and enterprise search.
- Moonshot — 1 model. Kimi K2.6 at $0.95 is the most capable reasoning model under $1/M input.
Compare All 53 Models Side by Side
Our comparison tool lets you filter by price, context window, provider, and capabilities across every tracked model.
Open Comparison Tool →The Bottom Line
The era of expensive AI is over. With 15 models under $1/M input tokens, every startup and indie developer can afford production-quality AI. The cheapest option, Gemini 2.5 Flash-Lite at $0.075/M, lets you process 13 million tokens for a dollar.
Here's the quick decision tree for choosing among these 15 models:
- Lowest possible cost → Llama 3.1 8B ($0.10/$0.10, total $0.20)
- Cheapest input for high volume → Gemini 2.5 Flash-Lite ($0.075 input)
- Best production chatbot → DeepSeek V4 Flash ($0.14/$0.28)
- Best budget coding → DeepSeek V4 Pro ($0.435/$0.87)
- No vendor lock-in → Llama 4 Scout ($0.18/$0.59, MIT license)
- EU data sovereignty → Mistral Small 4 ($0.15/$0.60)
- Best RAG on a budget → Command R ($0.50/$1.50)
- Best reasoning under $1 → Kimi K2.6 ($0.95/$4.00)
Use the APIpulse cost calculator to model your exact usage and find the cheapest model that meets your quality bar.
Stay ahead of API pricing changes
Get notified when providers change prices, deprecate models, or launch new ones. Join 2,400+ developers.
Stop guessing — get exact API cost comparisons
No signup required to 67-model comparison, migration code snippets, PDF reports, price alerts, and cost monitoring. ✅ All tools free.
Free Tools →