AI model APIs · cost calculator
What does your AI usage really cost?
AI APIs bill two meters at once — input tokens and output tokens — and the rates span 147× between the cheapest and priciest model at the same usage. Enter your monthly volume and every model is re-priced from its official rate card, cheapest first.
Lowest cost At 10M in / 3M out, go with Qwen-Flash — $1.70/mo cheapest frontier-class: GLM-5.2 at $27.2/mo
| Model | $/1M in · out | Cost / month | Link | |
|---|---|---|---|---|
| 01 |  Qwen-Flash Lowest cost | $0.05 · $0.40 | $1.70/mo | Details |
| 02 |  Gemini 2.5 Flash-Lite | $0.10 · $0.40 | $2.20/mo | Details |
| 03 |  DeepSeek V4 Flash | $0.14 · $0.28 | $2.24/mo | Details |
| 04 |  Mistral Small 4 | $0.15 · $0.60 | $3.30/mo | Details |
| 05 |  GPT-5.4 Nano | $0.20 · $1.25 | $5.75/mo | Details |
| 06 |  MiniMax M3 | $0.30 · $1.2 | $6.60/mo | Details |
| 07 | DeepSeek V4 Pro | $0.43 · $0.87 | $6.96/mo | Details |
| 08 | Gemini 3.1 Flash-Lite | $0.25 · $1.5 | $7/mo | Details |
| 09 | Qwen-Plus | $0.40 · $1.2 | $7.60/mo | Details |
| 10 | Mistral Large 3 | $0.50 · $1.5 | $9.50/mo | Details |
| 11 |  GLM-4.7 | $0.60 · $2.2 | $12.6/mo | Details |
| 12 |  Grok 4.3 | $1.25 · $2.5 | $20/mo | Details |
| 13 | GPT-5.4 Mini | $0.75 · $4.5 | $21/mo | Details |
| 14 |  Kimi K2.6 | $0.95 · $4 | $21.5/mo | Details |
| 15 |  Claude Haiku 4.5 | $1 · $5 | $25/mo | Details |
| 16 | GLM-5.2 | $1.4 · $4.4 | $27.2/mo | Details |
| 17 | GPT-5.6 Luna | $1 · $6 | $28/mo | Details |
| 18 | Qwen3-Max | $1.2 · $6 | $30/mo | Details |
| 19 | Mistral Medium 3.5 | $1.5 · $7.5 | $37.5/mo | Details |
| 20 | Grok 4.5 | $2 · $6 | $38/mo | Details |
| 21 | Gemini 3.5 Flash | $1.5 · $9 | $42/mo | Details |
| 22 | Gemini 3.1 Pro | $2 · $12 | $56/mo | Details |
| 23 | GPT-5.6 Terra | $2.5 · $15 | $70/mo | Details |
| 24 | Claude Sonnet 5 | $3 · $15 | $75/mo | Details |
| 25 | Kimi K3 | $3 · $15 | $75/mo | Details |
| 26 | Claude Opus 4.8 | $5 · $25 | $125/mo | Details |
| 27 | GPT-5.6 Sol | $5 · $30 | $140/mo | Details |
| 28 | Claude Fable 5 | $10 · $50 | $250/mo | Details |
Standard on-demand rates from each provider's official pricing page, verified July 2026. Figures exclude prompt-caching discounts, batch-API discounts (typically 50%), and long-context surcharges (Gemini 3.1 Pro and Grok roughly double above 200k-token prompts). Rankings are by price only — quality differs; confirm live pricing before committing.
Key facts · verified July 2026
- At 10M input / 3M output tokens a month, the same workload costs $1.70 on Qwen-Flash and $250 on Claude Fable 5 — a 147× spread.
- The cheapest frontier-class model is GLM-5.2 ($1.4 in / $4.4 out per 1M) — 9.2× cheaper than the priciest frontier model at the same usage.
- Output tokens cost 2× to 8× more than input on every model here (Qwen-Flash tops the ratio at 8×) — the input:output mix changes which model is cheapest, not just by how much.
Why your input:output mix matters
Per-token pricing is linear, so there are no volume cliffs — but because every provider marks up output tokens differently, the cheapest model flips with your mix. Two real examples from the current rate cards:
- GPT-5.6 Luna vs Grok 4.3: Luna is cheaper only while output makes up less than 7% of your tokens; past that, Grok 4.3's $2.50 output rate wins. See the full comparison →
- DeepSeek V4 Flash vs Gemini 2.5 Flash-Lite: Gemini wins input-heavy work, but once output passes 25% of tokens, DeepSeek takes over. See the full comparison →
All 28 models by provider
OpenAI
- GPT-5.6 Sol $5 / $30
- GPT-5.6 Terra $2.5 / $15
- GPT-5.6 Luna $1 / $6
- GPT-5.4 Mini $0.75 / $4.5
- GPT-5.4 Nano $0.20 / $1.25
Anthropic
- Claude Fable 5 $10 / $50
- Claude Opus 4.8 $5 / $25
- Claude Sonnet 5 $3 / $15
- Claude Haiku 4.5 $1 / $5
Google
- Gemini 3.1 Pro $2 / $12
- Gemini 3.5 Flash $1.5 / $9
- Gemini 3.1 Flash-Lite $0.25 / $1.5
- Gemini 2.5 Flash-Lite $0.10 / $0.40
DeepSeek
- DeepSeek V4 Pro $0.43 / $0.87
- DeepSeek V4 Flash $0.14 / $0.28
Mistral
- Mistral Medium 3.5 $1.5 / $7.5
- Mistral Large 3 $0.50 / $1.5
- Mistral Small 4 $0.15 / $0.60
MiniMax
- MiniMax M3 $0.30 / $1.2
Alibaba Qwen
- Qwen3-Max $1.2 / $6
- Qwen-Plus $0.40 / $1.2
- Qwen-Flash $0.05 / $0.40