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 |
|---|---|---|---|---|---|
| google/gemma-3-4b-itopenrouter/google/gemma-3-4b-it | openrouter | $0.05 | $0.10 | 131.1K | 16.4K |
| mistralai/mistral-small-24b-instruct-2501openrouter/mistralai/mistral-small-24b-instruct-2501 | openrouter | $0.05 | $0.08 | 32.8K | 16.4K |
| nvidia/nemotron-3-nano-30b-a3bnovita/nvidia/nemotron-3-nano-30b-a3b | novita | $0.05 | $0.20 | 262.1K | 32.8K |
| meta-llama/llama-3.2-3b-instructopenrouter/meta-llama/llama-3.2-3b-instruct | openrouter | $0.05 | $0.33 | 131.1K | 118K |
| accounts/fireworks/models/nemotron-lightning-3p5-30b-a3bfireworks_ai/accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b | Fireworks AI | $0.05 | $0.20 | 262.1K | 32.8K |
| inference-net/schematron-v2-smallopenrouter/inference-net/schematron-v2-small | openrouter | $0.05 | $0.23 | 128K | 4.1K |
| nemotron-lightning-3p5-30b-a3bfireworks_ai/nemotron-lightning-3p5-30b-a3b | Fireworks AI | $0.05 | $0.20 | 262.1K | 32.8K |
| qwen-turbo-2024-11-01qwencloud/qwen-turbo-2024-11-01 | qwencloud | $0.05 | $0.20 | 1M | 8.2K |
| qwen-turbo-latestqwencloud/qwen-turbo-latest | qwencloud | $0.05 | $0.20 | 1M | 16.4K |
| qwen-turboqwencloud/qwen-turbo | qwencloud | $0.05 | $0.20 | 129K | 16.4K |
| qwen-turbo-2025-04-28qwencloud/qwen-turbo-2025-04-28 | qwencloud | $0.05 | $0.20 | 1M | 16.4K |
| openai/gpt-6-luna-pro:batchopenrouter/openai/gpt-6-luna-pro:batch | openrouter | $0.05 | $0.25 | 1.1M | 128K |
| upstage/solar-mini4openrouter/upstage/solar-mini4 | openrouter | $0.05 | $0.20 | 524.3K | 131.1K |
| qwen-turbo-2024-11-01qwen_ai_platform/qwen-turbo-2024-11-01 | qwen_ai_platform | $0.05 | $0.20 | 1M | 8.2K |
| qwen-turbo-latestqwen_ai_platform/qwen-turbo-latest | qwen_ai_platform | $0.05 | $0.20 | 1M | 16.4K |
| openai/gpt-6-luna:batchopenrouter/openai/gpt-6-luna:batch | openrouter | $0.05 | $0.25 | 1.1M | 128K |
| qwen-turboqwen_ai_platform/qwen-turbo | qwen_ai_platform | $0.05 | $0.20 | 129K | 16.4K |
| qwen-turbo-2025-04-28qwen_ai_platform/qwen-turbo-2025-04-28 | qwen_ai_platform | $0.05 | $0.20 | 1M | 16.4K |
| Qwen/Qwen3.6-35B-A3Bsail/Qwen/Qwen3.6-35B-A3B | sail | $0.05 | $0.40 | 262.1K | 262.1K |
| anthropic/claude-haiku-5.5:batchopenrouter/anthropic/claude-haiku-5.5:batch | openrouter | $0.05 | $0.25 | 1M | 128K |
| summarizescaledown/summarize | scaledown | $0.05 | $0.00 | — | — |
| decisionsscaledown/decisions | scaledown | $0.05 | $0.00 | — | — |
| classifyscaledown/classify | scaledown | $0.05 | $0.00 | — | — |
| extractscaledown/extract | scaledown | $0.05 | $0.00 | — | — |
| compressscaledown/compress | scaledown | $0.05 | $0.00 | — | — |
| openai.gpt-5-nanooci/openai.gpt-5-nano | oci | $0.05 | $0.40 | 272K | 128K |
| llama-4-maverick-17b-128e-instruct-fp8lambda_ai/llama-4-maverick-17b-128e-instruct-fp8 | lambda_ai | $0.05 | $0.10 | 131.1K | 8.2K |
| llama-4-scout-17b-16e-instructlambda_ai/llama-4-scout-17b-16e-instruct | lambda_ai | $0.05 | $0.10 | 16.4K | 8.2K |
| qwen25-coder-32b-instructlambda_ai/qwen25-coder-32b-instruct | lambda_ai | $0.05 | $0.10 | 131.1K | 131.1K |
| qwen3-32b-fp8lambda_ai/qwen3-32b-fp8 | lambda_ai | $0.05 | $0.10 | 131.1K | 131.1K |
| qwen-turbo-2025-04-28dashscope/qwen-turbo-2025-04-28 | dashscope | $0.05 | $0.20 | 1M | 16.4K |
| qwen-turbo-latestdashscope/qwen-turbo-latest | dashscope | $0.05 | $0.20 | 1M | 16.4K |
| qwen-turbo-2024-11-01dashscope/qwen-turbo-2024-11-01 | dashscope | $0.05 | $0.20 | 1M | 8.2K |
| qwen-turbodashscope/qwen-turbo | dashscope | $0.05 | $0.20 | 129K | 16.4K |
| sao10k/l3-8b-lunarisnovita/sao10k/l3-8b-lunaris | novita | $0.05 | $0.05 | 8.2K | 8.2K |
| Sao10K/L3-8B-Stheno-v3.2novita/Sao10K/L3-8B-Stheno-v3.2 | novita | $0.05 | $0.05 | 8.2K | 32K |
| google/gemma-3-12b-itnovita/google/gemma-3-12b-it | novita | $0.05 | $0.10 | 131.1K | 8.2K |
| openai/gpt-oss-120bnovita/openai/gpt-oss-120b | novita | $0.05 | $0.25 | 131.1K | 32.8K |
| @cf/qwen/qwen3-30b-a3b-fp8cloudflare/@cf/qwen/qwen3-30b-a3b-fp8 | cloudflare | $0.05 | $0.34 | 32.8K | 32.8K |
| @cf/meta/llama-3.2-3b-instructcloudflare/@cf/meta/llama-3.2-3b-instruct | cloudflare | $0.05 | $0.34 | 80K | 80K |
| us/gpt-5-nanoazure/us/gpt-5-nano | Azure | $0.06 | $0.44 | — | — |
| eu/gpt-5-nanoazure/eu/gpt-5-nano | Azure | $0.06 | $0.44 | — | — |
| us/gpt-5-nano-2025-08-07azure/us/gpt-5-nano-2025-08-07 | Azure | $0.06 | $0.44 | 272K | 128K |
| eu/gpt-5-nano-2025-08-07azure/eu/gpt-5-nano-2025-08-07 | Azure | $0.06 | $0.44 | 272K | 128K |
| meta-llama/Llama-Guard-3-8Bdeepinfra/meta-llama/Llama-Guard-3-8B | DeepInfra | $0.06 | $0.06 | 131.1K | 131.1K |
| codellama-7bllamagate/codellama-7b | llamagate | $0.06 | $0.12 | 16.4K | 4.1K |
| qwen2.5-coder-7bllamagate/qwen2.5-coder-7b | llamagate | $0.06 | $0.12 | 32.8K | 8.2K |
| deepseek-coder-6.7bllamagate/deepseek-coder-6.7b | llamagate | $0.06 | $0.12 | 16.4K | 4.1K |
| amazon/nova-litevercel_ai_gateway/amazon/nova-lite | vercel_ai_gateway | $0.06 | $0.24 | 300K | 8.2K |
| nvidia.nemotron-nano-9b-v2nvidia.nemotron-nano-9b-v2 | bedrock_converse | $0.06 | $0.23 | 128K | 8K |
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 |
|---|---|---|
| google/gemma-3-4b-it | openrouter | $0.0200 |
| mistralai/mistral-small-24b-instruct-2501 | openrouter | $0.0180 |
| nvidia/nemotron-3-nano-30b-a3b | novita | $0.0300 |
| meta-llama/llama-3.2-3b-instruct | openrouter | $0.0430 |
| accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b | Fireworks AI | $0.0300 |
| inference-net/schematron-v2-small | openrouter | $0.0330 |
| nemotron-lightning-3p5-30b-a3b | Fireworks AI | $0.0300 |
| qwen-turbo-2024-11-01 | qwencloud | $0.0300 |
| qwen-turbo-latest | qwencloud | $0.0300 |
| qwen-turbo | qwencloud | $0.0300 |
| qwen-turbo-2025-04-28 | qwencloud | $0.0300 |
| openai/gpt-6-luna-pro:batch | openrouter | $0.0350 |
| upstage/solar-mini4 | openrouter | $0.0300 |
| qwen-turbo-2024-11-01 | qwen_ai_platform | $0.0300 |
| qwen-turbo-latest | qwen_ai_platform | $0.0300 |
| openai/gpt-6-luna:batch | openrouter | $0.0350 |
| qwen-turbo | qwen_ai_platform | $0.0300 |
| qwen-turbo-2025-04-28 | qwen_ai_platform | $0.0300 |
| Qwen/Qwen3.6-35B-A3B | sail | $0.0500 |
| anthropic/claude-haiku-5.5:batch | openrouter | $0.0350 |
| summarize | scaledown | $0.0100 |
| decisions | scaledown | $0.0100 |
| classify | scaledown | $0.0100 |
| extract | scaledown | $0.0100 |
| compress | scaledown | $0.0100 |
| openai.gpt-5-nano | oci | $0.0500 |
| llama-4-maverick-17b-128e-instruct-fp8 | lambda_ai | $0.0200 |
| llama-4-scout-17b-16e-instruct | lambda_ai | $0.0200 |
| qwen25-coder-32b-instruct | lambda_ai | $0.0200 |
| qwen3-32b-fp8 | lambda_ai | $0.0200 |