Cheapest AI API model (October 2026)
Sorted by blended price (3 input : 1 output tokens). Lowest right now: Claude Haiku 5.5 (Anthropic) at $0.10 in / $0.50 out per 1M tokens.
Cheapest from each provider: OpenAI: GPT-6 Luna ($0.20) · Anthropic: Claude Haiku 5.5 ($0.20) · Mistral: Mistral Small 4 ($0.2625) · DeepSeek: DeepSeek V4.1 Flash ($0.525) · Google: Gemini 3.5 Flash-Lite ($0.85) · xAI: Grok Build 0.1 ($1.25)
| Provider | Model | Input $/1M | Output $/1M | Cached input $/1M | Blended 3:1 | Context | Notes / source |
|---|---|---|---|---|---|---|---|
| Anthropic | Claude Haiku 5.5claude-haiku-5-5 | $0.10 | $0.50 | $0.01 | $0.20 | 1M | Prompts up to 100K tokens. Over 100K: $0.50 in / $2.50 out / $0.05 cached. Source, checked 2026-10-16 |
| OpenAI | GPT-6 Lunagpt-6-luna | $0.10 | $0.50 | $0.01 | $0.20 | 1.05M | Short-context rate. Long context: $0.20 in / $0.75 out. Source, checked 2026-10-16 |
| Mistral | Mistral Small 4mistral-small-4 | $0.15 | $0.60 | $0.015 | $0.2625 | Not listed | Context not listed on pricing page. Source, checked 2026-10-16 |
| Mistral | Codestralcodestral | $0.30 | $0.90 | $0.03 | $0.45 | Not listed | Context not listed on pricing page. Source, checked 2026-10-16 |
| DeepSeek | DeepSeek V4.1 Flashdeepseek-flash | $0.30 | $1.20 | $0.006 | $0.525 | 1M | Peak rate. Off-peak (most hours, weekends) is half: $0.15 in / $0.60 out. Source, checked 2026-10-16 |
Gemini 3.5 Flash-Litegemini-3.5-flash-lite | $0.30 | $2.50 | $0.03 | $0.85 | 1.05M | Paid tier. Free tier available. Source, checked 2026-10-16 | |
| Mistral | Mistral Large 4mistral-large-4 | $0.68 | $2.09 | $0.07 | $1.0325 | Not listed | Sale price (list $1.36 in / $4.18 out / $0.14 cached). Context not listed on pricing page. Source, checked 2026-10-16 |
| xAI | Grok Build 0.1grok-build-0.1 | $1.00 | $2.00 | $0.20 | $1.25 | 256K | Under 200K prompt tokens. At or above 200K: $2 in / $4 out. Source, checked 2026-10-16 |
Gemini 3.8 Flashgemini-3.8-flash | $0.75 | $3.75 | $0.075 | $1.50 | 1.05M | Paid tier. Price through Dec 31, 2026; from Jan 1, 2027: $1.50 in / $7.50 out. Free tier available. Source, checked 2026-10-16 | |
| xAI | Grok 4.3grok-4.3 | $1.25 | $2.50 | $0.20 | $1.5625 | 1M | Under 200K prompt tokens. At or above 200K: $2.50 in / $5 out. Source, checked 2026-10-16 |
| OpenAI | GPT-5.4 minigpt-5.4-mini | $0.75 | $4.50 | $0.075 | $1.6875 | 400K | Source, checked 2026-10-16 |
| DeepSeek | DeepSeek V4 Prodeepseek-v4-pro | $1.32 | $3.96 | $0.044 | $1.98 | 1M | Peak rate. Off-peak is half: $0.66 in / $1.98 out. Source, checked 2026-10-16 |
| xAI | Grok 4.7grok-4.7 | $2.00 | $6.00 | $0.50 | $3.00 | 500K | Under 200K prompt tokens. At or above 200K: $4 in / $12 out / $1 cached. Source, checked 2026-10-16 |
| Mistral | Mistral Medium 3.5mistral-medium-3.5 | $1.50 | $7.50 | $0.15 | $3.00 | Not listed | Context not listed on pricing page. Source, checked 2026-10-16 |
| Anthropic | Claude Sonnet 5.5claude-sonnet-5-5 | $2.00 | $10.00 | $0.10 | $4.00 | 1M | Cached = cache hits; 5m cache write $2.50, 1h write $4. Source, checked 2026-10-16 |
| OpenAI | GPT-6.1 Solgpt-6.1-sol | $2.00 | $10.00 | $0.10 | $4.00 | 1.05M | Short-context rate. Long context: $4 in / $15 out. Source, checked 2026-10-16 |
Gemini 4 Argongemini-4-argon | $2.00 | $10.00 | $0.10 | $4.00 | 1.05M | Paid tier. Price through Nov 30, 2026; from Dec 1, 2026: $4 in / $20 out / $0.20 cached. Not on free tier. Source, checked 2026-10-16 | |
| OpenAI | GPT-5.6 Terragpt-5.6-terra | $2.00 | $12.00 | $0.20 | $4.50 | 1.05M | Short-context rate. Long context: $4 in / $18 out. Source, checked 2026-10-16 |
| Anthropic | Claude Opus 5.5claude-opus-5-5 | $4.00 | $20.00 | $0.20 | $8.00 | 1M | Cached = cache hits; 5m cache write $5, 1h write $8. Source, checked 2026-10-16 |
| Anthropic | Claude Opus 4.8claude-opus-4-8 | $5.00 | $25.00 | $0.50 | $10.00 | Not listed | Older model; context not on current overview table. Source, checked 2026-10-16 |
| OpenAI | GPT-5.5gpt-5.5 | $5.00 | $30.00 | $0.50 | $11.25 | 1.05M | Under 272K context. Above: $10 in / $45 out. Source, checked 2026-10-16 |
| Anthropic | Claude Fable 5.1claude-fable-5-1 | $10.00 | $50.00 | $0.25 | $20.00 | 1M | Cached = cache hits; 5m cache write $12.50, 1h write $20. Source, checked 2026-10-16 |
| OpenAI | GPT-6 Astragpt-6-astra | $10.00 | $50.00 | $1.00 | $20.00 | 1.05M | Short-context rate. Long context: $20 in / $75 out. Source, checked 2026-10-16 |
Cheapest is not best for every job. See the full API price table for flagship models.