All Ultra-Budget Models Compared
The cheapest AI models available, ranked by input price.
| Model | Provider | Tier | Input (per 1M) | Output (per 1M) | Context |
|---|---|---|---|---|---|
| Gemini 2.5 Flash-Lite | Budget | $0.075 | $0.30 | 1M | |
| GPT-oss 20B | OpenAI | Budget | $0.08 | $0.35 | 128K |
| Gemini 2.5 Flash-Lite | Budget | $0.10 | $0.40 | 1M | |
| Llama 3.1 8B | Meta | Budget | $0.10 | $0.10 | 128K |
| DeepSeek V4 Flash | DeepSeek | Budget | $0.14 | $0.28 | 1M |
| GPT-oss 120B | OpenAI | Budget | $0.15 | $0.60 | 128K |
| GPT-4o mini | OpenAI | Budget | $0.15 | $0.60 | 128K |
| GPT-5 mini | OpenAI | Budget | $0.25 | $2.00 | 272K |
| Claude Haiku 4.5 | Anthropic | Budget | $1.00 | $5.00 | 200K |
Calculate Your Exact Costs
Pick your models, enter your usage, see which ultra-budget model saves you more.
Which Should You Choose?
Chatbot / Customer Support
High volume, short responses. Input tokens dominate. Cost per message matters most.
Content Generation
Long outputs, summarization, writing. Output tokens dominate. Cost per generation matters most.
RAG Pipeline / Classification
Large input contexts, short responses. Input-heavy workloads. Classification, extraction, tagging.
Code Generation
Mixed input/output. Longer outputs for code. Both handle most coding tasks well.
Long Document Analysis
Processing large documents with minimal output. Input-heavy. Both have 1M context.
Startup MVP / Side Project
Minimize costs while building. Need reliable API at the absolute lowest price point.
Save More with APIpulse
Get personalized cost optimization recommendations for your specific workload.
Frequently Asked Questions
Which is cheaper, DeepSeek V4 Flash or Gemini 2.5 Flash-Lite?
It depends on your usage. Gemini 2.5 Flash-Lite has the cheapest input at $0.075/1M tokens (vs DeepSeek's $0.14), but DeepSeek V4 Flash has the cheapest output at $0.28/1M tokens (vs Gemini's $0.30). For output-heavy workloads like content generation, DeepSeek is slightly cheaper. For input-heavy workloads like RAG or classification, Gemini Flash Lite wins.
What is the cheapest AI API model for input tokens?
Gemini 2.5 Flash-Lite at $0.075 per 1M tokens is the cheapest input pricing available. For comparison, DeepSeek V4 Flash costs $0.14, GPT-oss 20B costs $0.08, and GPT-5 mini costs $0.25. At 10M input tokens/month, Gemini Flash Lite costs $0.75 vs DeepSeek's $1.40.
What is the cheapest AI API model for output tokens?
DeepSeek V4 Flash at $0.28 per 1M tokens has the cheapest output pricing among major providers. Gemini 2.5 Flash-Lite costs $0.30, Gemini 2.5 Flash-Lite costs $0.40, and Llama 3.1 8B costs $0.10 (open-source via Together.ai). At 10M output tokens/month, DeepSeek costs $2.80 vs Gemini Flash Lite's $3.00.
DeepSeek V4 Flash vs Gemini Flash Lite for chatbots?
Both are excellent for chatbots. Gemini Flash Lite at $0.075/$0.30 with 1M context is ideal for high-volume, short-response chatbots. DeepSeek V4 Flash at $0.14/$0.28 with 1M context has slightly better output pricing. For most chatbot workloads, Gemini Flash Lite's cheaper input makes it the better choice since input tokens dominate in conversational use cases.
Are DeepSeek and Gemini Flash Lite available worldwide?
Gemini 2.5 Flash-Lite is available globally through Google AI Studio and Vertex AI. DeepSeek V4 Flash is available through DeepSeek's API and some third-party providers. Check APIpulse's provider pages for current availability and regional restrictions. Both models are in Budget tier, making them the cheapest options from their respective providers.
📊 Live Pricing
Real-time prices for all 67 models
💰 Pricing Hub
All 67 models compared — find the cheapest
Related Alternatives
All Tools Are Free
No signup required to 67-model comparison, migration code snippets, PDF reports, price alerts, and cost monitoring. ✅ All tools free.
Free Tools →