全球 AI 模型价格与规格对比
比较模型公开价格、上下文窗口与能力，按任务和提供商筛选。
模型列表
3564 个模型| 模型 | 提供商 | 输入 / 1M | 输出 / 1M | 上下文 | 最大输出 |
|---|---|---|---|---|---|
| apac.anthropic.claude-haiku-5-5apac.anthropic.claude-haiku-5-5 | bedrock_converse | $0.11 | $0.55 | 1M | 128K |
| jp.anthropic.claude-haiku-5-5jp.anthropic.claude-haiku-5-5 | bedrock_converse | $0.11 | $0.55 | 1M | 128K |
| anthropic.claude-haiku-5-5bedrock_mantle/anthropic.claude-haiku-5-5 | bedrock_mantle | $0.11 | $0.55 | 1M | 128K |
| meta-llama/llama-4-scout-17b-16e-instructgroq/meta-llama/llama-4-scout-17b-16e-instruct | Groq | $0.11 | $0.34 | 131.1K | 8.2K |
| us/gpt-4.1-nano-2025-04-14azure/us/gpt-4.1-nano-2025-04-14 | Azure | $0.11 | $0.44 | 1M | 32.8K |
| xiaomimimo/mimo-v2-flashnovita/xiaomimimo/mimo-v2-flash | novita | $0.11 | $0.33 | 262.1K | 32K |
| deepseek/deepseek-v4.1-flash:batchopenrouter/deepseek/deepseek-v4.1-flash:batch | openrouter | $0.11 | $0.34 | 1M | 131.1K |
| qwen3.5-122b-a10baihubmix/qwen3.5-122b-a10b | aihubmix | $0.11 | $0.90 | 262.1K | 65.5K |
| qwen3.8-flashaihubmix/qwen3.8-flash | aihubmix | $0.11 | $0.38 | 1M | 131.1K |
| glm-5.3-flashaihubmix/glm-5.3-flash | aihubmix | $0.11 | $0.39 | 1M | 131.1K |
| qwen/qwen3-8bopenrouter/qwen/qwen3-8b | openrouter | $0.12 | $0.46 | 131.1K | 8.2K |
| qwen/qwen3-vl-8b-instructopenrouter/qwen/qwen3-vl-8b-instruct | openrouter | $0.12 | $0.46 | 262.1K | 32.8K |
| google/gemma-3-27b-itnovita/google/gemma-3-27b-it | novita | $0.12 | $0.20 | 98.3K | 16.4K |
| llama3.1-70b-instruct-fp8lambda_ai/llama3.1-70b-instruct-fp8 | lambda_ai | $0.12 | $0.30 | 131.1K | 131.1K |
| llama3.3-70b-instruct-fp8lambda_ai/llama3.3-70b-instruct-fp8 | lambda_ai | $0.12 | $0.30 | 131.1K | 131.1K |
| llama3.1-nemotron-70b-instruct-fp8lambda_ai/llama3.1-nemotron-70b-instruct-fp8 | lambda_ai | $0.12 | $0.30 | 131.1K | 131.1K |
| hermes3-70blambda_ai/hermes3-70b | lambda_ai | $0.12 | $0.30 | 131.1K | 131.1K |
| qwen/qwen3-30b-a3bopenrouter/qwen/qwen3-30b-a3b | openrouter | $0.12 | $0.50 | 131.1K | 16.4K |
| qwen/qwen3-coder-nextopenrouter/qwen/qwen3-coder-next | openrouter | $0.12 | $0.80 | 262.1K | 235.9K |
| qwen/qwen3-14bopenrouter/qwen/qwen3-14b | openrouter | $0.12 | $0.24 | 131.1K | 16.4K |
| eu/gpt-6-lunaazure/eu/gpt-6-luna | Azure | $0.12 | $0.60 | 922K | 128K |
| us-gov-west-1/anthropic.claude-haiku-5-5bedrock/us-gov-west-1/anthropic.claude-haiku-5-5 | Amazon | $0.12 | $0.60 | 1M | 128K |
| us-gov-east-1/anthropic.claude-haiku-5-5bedrock/us-gov-east-1/anthropic.claude-haiku-5-5 | Amazon | $0.12 | $0.60 | 1M | 128K |
| us-gov.anthropic.claude-haiku-5-5us-gov.anthropic.claude-haiku-5-5 | bedrock_converse | $0.12 | $0.60 | 1M | 128K |
| us-gov-west-1/anthropic.claude-haiku-5-5bedrock_mantle/us-gov-west-1/anthropic.claude-haiku-5-5 | bedrock_mantle | $0.12 | $0.60 | 1M | 128K |
| cohere/embed-v4.0vercel_ai_gateway/cohere/embed-v4.0 | vercel_ai_gateway | $0.12 | $0.00 | — | — |
| Qwen/Qwen3-30B-A3Bdeepinfra/Qwen/Qwen3-30B-A3B | DeepInfra | $0.12 | $0.50 | 41K | 41K |
| Qwen/Qwen3-14Bdeepinfra/Qwen/Qwen3-14B | DeepInfra | $0.12 | $0.24 | 41K | 41K |
| Qwen/Qwen2.5-72B-Instructhyperbolic/Qwen/Qwen2.5-72B-Instruct | hyperbolic | $0.12 | $0.30 | 131.1K | 131.1K |
| meta-llama/Meta-Llama-3.1-70B-Instructhyperbolic/meta-llama/Meta-Llama-3.1-70B-Instruct | hyperbolic | $0.12 | $0.30 | 32.8K | 32.8K |
| meta-llama/Meta-Llama-3-70B-Instructhyperbolic/meta-llama/Meta-Llama-3-70B-Instruct | hyperbolic | $0.12 | $0.30 | 131.1K | 131.1K |
| meta-llama/Meta-Llama-3.1-405B-Instructhyperbolic/meta-llama/Meta-Llama-3.1-405B-Instruct | hyperbolic | $0.12 | $0.30 | 32.8K | 32.8K |
| meta-llama/Meta-Llama-3.1-8B-Instructhyperbolic/meta-llama/Meta-Llama-3.1-8B-Instruct | hyperbolic | $0.12 | $0.30 | 32.8K | 32.8K |
| meta-llama/Llama-3.3-70B-Instructhyperbolic/meta-llama/Llama-3.3-70B-Instruct | hyperbolic | $0.12 | $0.30 | 131.1K | 131.1K |
| Qwen/Qwen2.5-Coder-32B-Instructhyperbolic/Qwen/Qwen2.5-Coder-32B-Instruct | hyperbolic | $0.12 | $0.30 | 32.8K | 32.8K |
| NousResearch/Hermes-3-Llama-3.1-70Bhyperbolic/NousResearch/Hermes-3-Llama-3.1-70B | hyperbolic | $0.12 | $0.30 | 32.8K | 32.8K |
| meta-llama/Llama-3.2-3B-Instructhyperbolic/meta-llama/Llama-3.2-3B-Instruct | hyperbolic | $0.12 | $0.30 | 32.8K | 32.8K |
| chat-bison-001palm/chat-bison-001 | $0.13 | $0.13 | 8.2K | 4.1K | |
| chat-bisonpalm/chat-bison | $0.13 | $0.13 | 8.2K | 4.1K | |
| google/gemini-3.1-flash-lite:batchopenrouter/google/gemini-3.1-flash-lite:batch | openrouter | $0.13 | $0.75 | 1M | 65.5K |
| openai/gpt-5-mini:batchopenrouter/openai/gpt-5-mini:batch | openrouter | $0.13 | $1.00 | 400K | 128K |
| Phi-4azure_ai/Phi-4 | Azure | $0.13 | $0.50 | 16.4K | 16.4K |
| Phi-4-reasoningazure_ai/Phi-4-reasoning | Azure | $0.13 | $0.50 | 32.8K | 4.1K |
| ps/glm-4.5-airpinstripes/ps/glm-4.5-air | pinstripes | $0.13 | $0.45 | 128K | 128K |
| eu.meta.llama3-2-1b-instruct-v1:0eu.meta.llama3-2-1b-instruct-v1:0 | Amazon | $0.13 | $0.13 | 128K | 4.1K |
| zai-org/glm-4.5-airnovita/zai-org/glm-4.5-air | novita | $0.13 | $0.85 | 131.1K | 98.3K |
| Phi-3.5-mini-instructazure_ai/Phi-3.5-mini-instruct | Azure | $0.13 | $0.52 | 128K | 4.1K |
| Phi-3-mini-128k-instructazure_ai/Phi-3-mini-128k-instruct | Azure | $0.13 | $0.52 | 128K | 4.1K |
| Phi-3-mini-4k-instructazure_ai/Phi-3-mini-4k-instruct | Azure | $0.13 | $0.52 | 4.1K | 4.1K |
| Phi-3.5-vision-instructazure_ai/Phi-3.5-vision-instruct | Azure | $0.13 | $0.52 | 128K | 4.1K |
估算调用费用
按公开单价估算。阶梯价、缓存、批量折扣和税费可能改变实际账单。
| 模型 | 提供商 | 预计费用 |
|---|---|---|
| apac.anthropic.claude-haiku-5-5 | bedrock_converse | $0.0770 |
| jp.anthropic.claude-haiku-5-5 | bedrock_converse | $0.0770 |
| anthropic.claude-haiku-5-5 | bedrock_mantle | $0.0770 |
| meta-llama/llama-4-scout-17b-16e-instruct | Groq | $0.0560 |
| us/gpt-4.1-nano-2025-04-14 | Azure | $0.0660 |
| xiaomimimo/mimo-v2-flash | novita | $0.0550 |
| deepseek/deepseek-v4.1-flash:batch | openrouter | $0.0560 |
| qwen3.5-122b-a10b | aihubmix | $0.1126 |
| qwen3.8-flash | aihubmix | $0.0605 |
| glm-5.3-flash | aihubmix | $0.0620 |
| qwen/qwen3-8b | openrouter | $0.0689 |
| qwen/qwen3-vl-8b-instruct | openrouter | $0.0689 |
| google/gemma-3-27b-it | novita | $0.0438 |
| llama3.1-70b-instruct-fp8 | lambda_ai | $0.0540 |
| llama3.3-70b-instruct-fp8 | lambda_ai | $0.0540 |
| llama3.1-nemotron-70b-instruct-fp8 | lambda_ai | $0.0540 |
| hermes3-70b | lambda_ai | $0.0540 |
| qwen/qwen3-30b-a3b | openrouter | $0.0740 |
| qwen/qwen3-coder-next | openrouter | $0.1040 |
| qwen/qwen3-14b | openrouter | $0.0480 |
| eu/gpt-6-luna | Azure | $0.0840 |
| us-gov-west-1/anthropic.claude-haiku-5-5 | Amazon | $0.0840 |
| us-gov-east-1/anthropic.claude-haiku-5-5 | Amazon | $0.0840 |
| us-gov.anthropic.claude-haiku-5-5 | bedrock_converse | $0.0840 |
| us-gov-west-1/anthropic.claude-haiku-5-5 | bedrock_mantle | $0.0840 |
| cohere/embed-v4.0 | vercel_ai_gateway | $0.0240 |
| Qwen/Qwen3-30B-A3B | DeepInfra | $0.0740 |
| Qwen/Qwen3-14B | DeepInfra | $0.0480 |
| Qwen/Qwen2.5-72B-Instruct | hyperbolic | $0.0540 |
| meta-llama/Meta-Llama-3.1-70B-Instruct | hyperbolic | $0.0540 |