Models and pricing
Prefer high-quality models. Better models produce better work — fewer iterations, fewer mistakes, less time spent reviewing. The cost difference is usually worth it. Recommended models are marked below.
Effective costs are lower than they look. Most tokens are cached input, not fresh input or output. Our Claude usage has a ~90% cache hit rate, and output is less than 1% of total tokens. At those ratios, Claude Opus 4.6 costs roughly $1/M tokens — not the $5.00 input price.
Prices are in US dollars per million tokens.
| Model | Input | Input cache | Output |
|---|---|---|---|
| OpenAI | |||
| GPT-4.1 | |||
gpt-4.1 (gpt-4.1-2025-04-14) |
$2.00 | $0.50 | $8.00 |
gpt-4.1-mini (gpt-4.1-mini-2025-04-14) |
$0.40 | $0.10 | $1.60 |
gpt-4.1-nano (gpt-4.1-nano-2025-04-14) |
$0.10 | $0.03 | $0.40 |
| GPT-5 | |||
gpt-5 (gpt-5-2025-08-07) |
$1.25 | $0.12 | $10.00 |
gpt-5-mini (gpt-5-mini-2025-08-07) |
$0.25 | $0.03 | $2.00 |
gpt-5-codex |
$1.25 | $0.12 | $10.00 |
| GPT-5.1 | |||
gpt-5.1 (gpt-5.1-2025-11-13) |
$1.25 | $0.12 | $10.00 |
gpt-5.1-codex |
$1.25 | $0.12 | $10.00 |
gpt-5.1-codex-mini |
$0.25 | $0.03 | $2.00 |
| GPT-5.2 | |||
gpt-5.2 (gpt-5.2-2025-12-11) |
$1.75 | $0.17 | $14.00 |
gpt-5.2-codex Recommended |
$1.75 | $0.17 | $14.00 |
| GPT-5.3 | |||
gpt-5.3-codex |
$1.75 | $0.17 | $14.00 |
| Mistral | |||
| Devstral 2 | |||
devstral-latest Recommended |
$0.40 | — | $2.00 |
labs-devstral-small-2512 |
$0.10 | — | $0.30 |
| Mistral (Vertex) | |||
codestral-2 |
$0.30 | — | $0.90 |
mistral-medium-3 |
$0.40 | — | $2.00 |
mistral-small-2503 |
$0.10 | — | $0.30 |
| Vertex AI | |||
| Claude | |||
claude-haiku-4-5 |
$1.00 | $0.10 read, $1.25 write | $5.00 |
claude-sonnet-4-5 |
$3.00 | $0.30 read, $3.75 write | $15.00 |
claude-sonnet-4-6 |
$3.00 | $0.30 read, $3.75 write | $15.00 |
claude-opus-4-5 |
$5.00 | $0.50 read, $6.25 write | $25.00 |
claude-opus-4-6 Recommended |
$5.00 | $0.50 read, $6.25 write | $25.00 |
| Gemini 2.5 | |||
gemini-2.5-pro |
$1.25 | $0.12 | $10.00 |
gemini-2.5-flash |
$0.30 | — | $2.50 |
| Gemini 3 | |||
gemini-3-pro-preview |
$2.00 | $0.20 | $12.00 |
gemini-3-flash-preview |
$0.50 | $0.05 | $3.00 |
| Gemini 3.1 | |||
gemini-3.1-pro-preview Recommended |
$2.00 | $0.20 | $12.00 |
| DeepSeek | |||
deepseek-ai/deepseek-r1-0528-maas |
$1.35 | — | $5.40 |
deepseek-ai/deepseek-v3.1-maas |
$0.60 | — | $1.70 |
deepseek-ai/deepseek-v3.2-maas |
$0.56 | $0.06 | $1.68 |
| Llama | |||
meta/llama-3.3-70b-instruct-maas |
$0.72 | — | $0.72 |
meta/llama-4-maverick-17b-128e-instruct-maas |
$0.35 | — | $1.15 |
| Qwen | |||
qwen/qwen3-235b-a22b-instruct-2507-maas |
$0.25 | — | $1.00 |
qwen/qwen3-coder-480b-a35b-instruct-maas |
$1.00 | — | $4.00 |
| Kimi | |||
moonshotai/kimi-k2-thinking-maas |
$0.60 | — | $2.50 |
| MiniMax | |||
minimaxai/minimax-m2-maas |
$0.30 | — | $1.20 |
| OpenAI OSS | |||
openai/gpt-oss-120b-maas |
$0.09 | — | $0.36 |
openai/gpt-oss-20b-maas |
$0.07 | — | $0.25 |
| GLM | |||
zai-org/glm-4.7-maas |
$0.60 | — | $2.20 |
zai-org/glm-5-maas Recommended |
$1.00 | $0.10 | $3.20 |