Prices are per 1M tokens. All data verified Jul 9, 2026. Full pricing table →
| # | Model | Provider | Input | Output | Context |
|---|---|---|---|---|---|
| 1 | Gemini 2.5 Flash-Lite | $0.075 | $0.30 | 1M | |
| 2 | GPT-oss 20B | OpenAI | $0.08 | $0.35 | 128K |
| 3 | Gemini 2.5 Flash-Lite | $0.10 | $0.40 | 1M | |
| 4 | Llama 3.1 8B | Meta (Together.ai) | $0.10 | $0.10 | 128K |
| 5 | Llama 4 Scout | Meta (Together.ai) | $0.11 | $0.34 | 10M |
| 6 | DeepSeek V4 Flash | DeepSeek | $0.14 | $0.28 | 1M |
| 7 | GPT-4o mini | OpenAI | $0.15 | $0.60 | 128K |
| 8 | GPT-oss 120B | OpenAI | $0.15 | $0.60 | 128K |
| 9 | Mistral Small 4 | Mistral | $0.15 | $0.60 | 128K |
| 10 | Llama 4 Maverick | Meta (Together.ai) | $0.20 | $0.60 | 10M |
| 11 | GPT-5 mini | OpenAI | $0.25 | $2.00 | 272K |
| 12 | DeepSeek V3 | DeepSeek | $0.27 | $1.10 | 128K |
| 13 | Mistral Large 3 | Mistral | $0.50 | $1.50 | 128K |
| 14 | Command R | Cohere | $0.50 | $1.50 | 128K |
| 15 | DeepSeek V4 Pro | DeepSeek | $0.44 | $0.87 | 1M |
| 16 | Kimi K2.6 | Moonshot | $0.90 | $3.75 | 256K |
| 17 | Claude Haiku 4.5 | Anthropic | $1.00 | $5.00 | 200K |
| 18 | Gemini 2.5 Pro | $1.25 | $10.00 | 1M | |
| 19 | GPT-5 | OpenAI | $1.25 | $10.00 | 272K |
| 20 | GPT-5.3 Codex | OpenAI | $1.75 | $14.00 | 400K |
| 21 | Gemini 3.1 Pro | $2.00 | $12.00 | 1M | |
| 22 | Jamba 1.5 Large | AI21 | $2.00 | $8.00 | 256K |
| 23 | GPT-4o | OpenAI | $2.50 | $10.00 | 128K |
| 24 | Command R+ | Cohere | $2.50 | $10.00 | 128K |
| 25 | Claude Sonnet 4.6 | Anthropic | $3.00 | $15.00 | 200K |
| 26 | Claude Sonnet 4.6 | Anthropic | $3.00 | $15.00 | 1M |
| 27 | Grok Build 0.1 | xAI | $0.30 | $0.50 | 256K |
| 28 | GPT-5.5 | OpenAI | $5.00 | $30.00 | 1M |
| 29 | Claude Opus 4.7 | Anthropic | $5.00 | $25.00 | 1M |
| 30 | Claude Opus 4.8 | Anthropic | $5.00 | $25.00 | 1M |
| 31 | Llama 3.1 70B | Meta (Together.ai) | $0.88 | $0.88 | 128K |
| 32 | Claude 4 Opus | Anthropic | $15.00 | $75.00 | 200K |
| 33 | Grok 4.3 | xAI | $1.25 | $2.50 | 1M |
| 34 | GPT-5.5 Pro | OpenAI | $30.00 | $180.00 | 1M |
Real Cost Scenarios
Let's see what these prices actually mean for real workloads. We'll model three common use cases.
Scenario 1: Simple Chatbot (10K requests/day)
A customer support chatbot with ~500 input tokens and ~300 output tokens per request.
Monthly Cost Comparison
Scenario 2: Code Assistant (5K requests/day)
A coding assistant with ~2,000 input tokens and ~4,000 output tokens per request (longer outputs).
Monthly Cost Comparison
Scenario 3: RAG Pipeline (1K requests/day)
A RAG system with ~4,000 input tokens (context + query) and ~1,000 output tokens.
Monthly Cost Comparison
Best Budget Models by Use Case
Best for Simple Tasks (Classification, Extraction, Formatting)
Gemini 2.5 Flash-Lite ($0.075/$0.30) or Llama 3.1 8B ($0.10/$0.10). These models handle straightforward tasks with minimal quality loss at 90%+ savings vs premium models.
Best for Code Generation
DeepSeek V4 Pro ($0.44/$0.87). Best price-to-quality ratio for coding. 1M context window handles large codebases. For simpler completions, Mistral Small 4 ($0.15/$0.60)) is a strong budget pick.
Best for Long Context
Llama 4 Scout (1M context) or Gemini 2.5 Flash-Lite ($0.10/$0.40) with 1M context. Both are extremely affordable for long-document processing.
Best for RAG Pipelines
DeepSeek V4 Flash ($0.14/$0.28). 1M context window, very low output cost. For RAG with shorter contexts, GPT-4o mini ($0.15/$0.60) is also competitive.
Best Quality-per-Dollar (Mid-Tier)
DeepSeek V4 Pro ($0.44/$0.87) or Mistral Large 3 ($0.50/$1.50). Both offer strong reasoning at a fraction of GPT-5/Claude Sonnet pricing.
How to Choose the Right Model
The cheapest model isn't always the best choice. Here's a decision framework:
- Start with the task complexity. Simple tasks (classification, extraction, formatting) → budget models. Complex reasoning, code generation, creative writing → mid-tier or premium.
- Consider output length. If your outputs are long (code, analysis), output pricing matters more. DeepSeek and Gemini have the lowest output costs.
- Check context window needs. If you're processing long documents, you need 1M+ context. Gemini Flash and Llama 4 Scout are the cheapest with large contexts.
- Test before committing. Use our Prompt Cost Calculator to model your exact workload across all 67 models.
Calculate Your Exact Costs
Paste your actual prompt and see costs across all 67 models. Find the cheapest option for your specific workload.
Open Calculator — Free🎯 Rate Your API Setup in 30 Seconds
Get an A+ to F grade on your AI API costs. See how you compare and find cheaper alternatives instantly.
Get Your Cost Score →📊 Generate Your Personalized API Cost Report
Select your model, enter your monthly spend, and get a custom savings report with cheaper alternatives — free, in 60 seconds.
Key Takeaways
- GPT-oss 20B is the cheapest at $0.08/$0.35 per 1M tokens
- Budget models are 30-400x cheaper than premium models for most tasks
- DeepSeek V4 Pro offers the best price-to-quality ratio for complex tasks
- Output pricing matters more than input pricing for most use cases
- Test with real prompts — use our calculator to find the sweet spot for your workload
Methodology
All prices sourced directly from provider pricing pages, verified Jul 9, 2026. Prices are per 1M tokens. We track 67 models across 10 providers: OpenAI, Anthropic, Google, DeepSeek, Mistral, Cohere, Meta (Together.ai), Moonshot, xAI, and AI21. Data is updated monthly. See pricing changelog →
Compare models: APIpulse Model Comparison — side-by-side pricing for 67 models across 10 providers. Free tool.
Want to optimize your AI API costs?
APIpulse includes free cost comparisons, exports, and recommendations that can save you up to 40%.
Free Cost Audit →