Pricing guide
Cheapest AI Models That Still Hold Up
Cheap is only useful if the model is still competent. These are the low-cost options that still deserve production trials.
| Model | Input | Output | Total price signal | Best for |
|---|---|---|---|---|
| GPT-OSS-120B OpenAI | $0 | $0 | $0.00 | Self-hosted |
| DeepSeek V4 Flash DeepSeek | $0.14 | $0.28 | $0.42 | Low-cost agent experiments |
| DeepSeek V4 Pro DeepSeek | $0.44 | $1.32 | $1.76 | Budget coding |
| GPT-5.6 Luna OpenAI | $0.4 | $1.8 | $2.20 | High-volume workflows |
| Gemini 3.5 Flash-Lite | $0.3 | $2.5 | $2.80 | High-volume automation |
| GLM-5 Zhipu AI | $1 | $3.2 | $4.20 | Bilingual (CN/EN) |
| Gemini 3.7 Flash | $0.75 | $3.75 | $4.50 | Agentic coding |
| GLM-5.3 Z.ai | $1.4 | $4.4 | $5.80 | Complex software engineering |
| Claude Haiku 4.5 Anthropic | $1 | $5 | $6.00 | Fast responses |
| Grok 4.5 xAI | $2 | $6 | $8.00 | Agentic coding |
| Grok 4.6 xAI | $2 | $6 | $8.00 | Agentic coding |
| Gemini 3.6 Flash | $1.5 | $7.5 | $9.00 | Agentic coding |
| Mistral Medium 3.5 Mistral | $1.5 | $7.5 | $9.00 | European compliance |
| Gemini 3.5 Flash | $1.5 | $9 | $10.50 | Fast multimodal agents |
| Claude Sonnet 5 Anthropic | $2 | $10 | $12.00 | Balanced performance |
| GPT-5.2-Codex OpenAI | $1.75 | $14 | $15.75 | Coding-focused tasks |
| GPT-5.2 OpenAI | $1.75 | $14 | $15.75 | General-purpose |
| GPT-5.4 OpenAI | $2.5 | $15 | $17.50 | Coding |
| Kimi K3 Moonshot AI | $3 | $15 | $18.00 | Long-horizon coding |
| GPT-5.6 Terra OpenAI | $4 | $18 | $22.00 | Production agents |
| Claude Opus 5 Anthropic | $5 | $25 | $30.00 | Complex reasoning |
| Claude Opus 4.8 Anthropic | $5 | $25 | $30.00 | Complex reasoning |
| GPT-5.5 OpenAI | $5 | $30 | $35.00 | Complex reasoning |
| GPT-5.6 Sol OpenAI | $10 | $45 | $55.00 | Complex production workflows |
| Claude Fable 5 Anthropic | $10 | $50 | $60.00 | Long-running agents |
Best ultra-budget option
DeepSeek V4 Pro combines low listed API pricing, 1M context, and hybrid thinking modes for cost-sensitive production trials.
Best value frontier model
MiniMax M2.5 is the best balance of serious quality and low token cost.
Best bilingual value
GLM-5 is still one of the strongest value models if Chinese/English matters.
When cheap is the right choice
- High-volume automations
- Background summarization and extraction
- Internal tooling where cost beats polish
- Early-stage startups watching burn