AI model guide
API prices on this page are USD only (no official CAD price exists), including any prices in the notes. A short guide to each provider's current flagship models, using the API prices and plan data on this site. This is a draft: model descriptions are limited to what providers state on their pricing and docs pages.
Want to know which model scores best? See Model rankings from independent benchmarks.
OpenAI
- gpt-6-astra: API: US$10.00 in / US$50.00 out per 1M tokens (USD only; no official CAD price). Short ctx ≤272K; long ctx $20/$75; batch/flex 50% off (platform.openai.com, 2026-10-14)
- gpt-6.1-sol: API: US$2.00 in / US$10.00 out per 1M tokens (USD only; no official CAD price). Long ctx $4/$15 (platform.openai.com, 2026-10-14)
- gpt-6-luna: API: US$0.10 in / US$0.50 out per 1M tokens (USD only; no official CAD price). Cheapest GPT-6 (platform.openai.com, 2026-10-14)
- gpt-5.6-sol: API: US$4.00 in / US$20.00 out per 1M tokens (USD only; no official CAD price). Promotional pricing through at least 2026-11-21 (platform.openai.com, 2026-10-14)
Consumer plans: OpenAI plans.
Anthropic
- Fable 5.1: API: US$10 in / US$50 out per 1M tokens (USD only; no official CAD price), context Up to 1M (varies by model). Cache write $12.50; US-only inference 1.1x (claude.com, 2026-10-14)
- Opus 5.5: API: US$4 in / US$20 out per 1M tokens (USD only; no official CAD price), context Up to 1M (varies by model). Cache write $5 (claude.com, 2026-10-14)
- Sonnet 5.5: API: US$2 in / US$10 out per 1M tokens (USD only; no official CAD price), context Up to 1M (varies by model). Cache write $2.50; cache read not clearly shown (claude.com, 2026-10-14)
- Haiku 5.5: API: US$0.50 in / US$2.50 out per 1M tokens (USD only; no official CAD price), context Up to 1M (varies by model). Prompts ≤100K; cached figures ($0.125/$0.625) ambiguous in page (claude.com, 2026-10-14)
Consumer plans: Anthropic plans.
- gemini-4-argon: API: US$2.00 in / US$10.00 out per 1M tokens (USD only; no official CAD price), context 1M. Promo through 2026-11-30; doubles to $4/$20 (cache $0.20) from 2026-12-01 (ai.google.dev, 2026-10-14)
- gemini-3.8-flash: API: US$0.75 in / US$3.75 out per 1M tokens (USD only; no official CAD price). Promo through 2026-12-31; $1.50/$7.50 from 2027-01-01; free tier available (ai.google.dev, 2026-10-14)
- gemini-3.1-pro-preview: API: US$2.00 in / US$12.00 out per 1M tokens (USD only; no official CAD price). ≤200K prompts; >200K $4/$18 (ai.google.dev, 2026-10-14)
- gemini-3.5-flash-lite: API: US$0.30 in / US$2.50 out per 1M tokens (USD only; no official CAD price). (ai.google.dev, 2026-10-14)
Consumer plans: Google plans.
xAI
- grok-4.7: API: US$2.00 in / US$6.00 out per 1M tokens (USD only; no official CAD price), context 500K. <200K prompt; ≥200K $4/$12 (docs.x.ai, 2026-10-14)
- grok-4.3: API: US$1.25 in / US$2.50 out per 1M tokens (USD only; no official CAD price), context 1M. ≥200K $2.50/$5 (docs.x.ai, 2026-10-14)
Consumer plans: xAI plans.
Meta
- Llama API: price unverified. NOT VERIFIED (llama.developer.meta.com, 2026-10-14)
Consumer plans: Meta plans.
Mistral
- Mistral Large 4: API: US$0.68 in / US$2.09 out per 1M tokens (USD only; no official CAD price). Sale price (orig $1.36/$4.18) (docs.mistral.ai, 2026-10-14)
- Mistral Medium 3.5: API: US$1.5 in / US$7.5 out per 1M tokens (USD only; no official CAD price), context 256K. (docs.mistral.ai, 2026-10-14)
- Mistral Small 4: API: US$0.15 in / US$0.6 out per 1M tokens (USD only; no official CAD price). (docs.mistral.ai, 2026-10-14)
Consumer plans: Mistral plans.
Perplexity
- Sonar API: price unverified. NOT VERIFIED (docs not fetched) (docs.perplexity.ai, 2026-10-14)
Consumer plans: Perplexity plans.
DeepSeek
- deepseek-v4-pro: API: US$1.32 in / US$3.96 out per 1M tokens (USD only; no official CAD price), context 1M. Peak rates; off-peak half ($0.66/$1.98/$0.022) (api-docs.deepseek.com, 2026-10-14)
- deepseek-flash (V4.1-Flash): API: US$0.30 in / US$1.20 out per 1M tokens (USD only; no official CAD price), context 1M. Peak; off-peak $0.15/$0.60/$0.003 (api-docs.deepseek.com, 2026-10-14)
Consumer plans: DeepSeek plans.