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
3574 个模型| 模型 | 提供商 | 输入 / 1M | 输出 / 1M | 上下文 | 最大输出 |
|---|---|---|---|---|---|
| meta-llama/Llama-3-8b-chat-hftogether_ai/meta-llama/Llama-3-8b-chat-hf | together_ai | $0.20 | $0.20 | 8.2K | — |
| coding-xiaomi-mimo-v2.5-proaihubmix/coding-xiaomi-mimo-v2.5-pro | aihubmix | $0.20 | $0.40 | 1M | 131.1K |
| gpt-5.6-lunaaihubmix/gpt-5.6-luna | aihubmix | $0.20 | $1.20 | 1.1M | 128K |
| gpt-5.4-nanoaihubmix/gpt-5.4-nano | aihubmix | $0.20 | $1.25 | 400K | 128K |
| gpt-5.6-luna-2026-07-09azure/gpt-5.6-luna-2026-07-09 | Azure | $0.20 | $1.20 | 922K | 128K |
| stepfun/step-3.7-flashopenrouter/stepfun/step-3.7-flash | openrouter | $0.20 | $1.15 | 262.1K | 230.4K |
| openai/gpt-4.1-mini:batchopenrouter/openai/gpt-4.1-mini:batch | openrouter | $0.20 | $0.80 | 1M | 32.8K |
| mistralai/mistral-medium-3.1:batchopenrouter/mistralai/mistral-medium-3.1:batch | openrouter | $0.20 | $1.00 | 131.1K | 104.9K |
| cognitivecomputations/dolphin-mistral-24b-venice-editionopenrouter/cognitivecomputations/dolphin-mistral-24b-venice-edition | openrouter | $0.20 | $0.90 | 128K | 8.2K |
| llama-guard-3-8bgroq/llama-guard-3-8b | Groq | $0.20 | $0.20 | 8.2K | — |
| llama-3.1-8b-instructperplexity/llama-3.1-8b-instruct | Perplexity | $0.20 | $0.20 | 131.1K | 131.1K |
| zai/glm-4.5-airvercel_ai_gateway/zai/glm-4.5-air | vercel_ai_gateway | $0.20 | $1.10 | 128K | 96K |
| alibaba/qwen-3-235bvercel_ai_gateway/alibaba/qwen-3-235b | vercel_ai_gateway | $0.20 | $0.60 | 41K | 16.4K |
| meta/llama-4-maverickvercel_ai_gateway/meta/llama-4-maverick | vercel_ai_gateway | $0.20 | $0.60 | 131.1K | 8.2K |
| google/gemma-2-9bvercel_ai_gateway/google/gemma-2-9b | vercel_ai_gateway | $0.20 | $0.20 | 8.2K | 8.2K |
| mistralai/ministral-14b-2512openrouter/mistralai/ministral-14b-2512 | openrouter | $0.20 | $0.20 | 262.1K | 209.7K |
| deepseek-ai/DeepSeek-R1-Distill-Llama-70Bdeepinfra/deepseek-ai/DeepSeek-R1-Distill-Llama-70B | DeepInfra | $0.20 | $0.60 | 131.1K | 131.1K |
| Qwen/Qwen2.5-VL-32B-Instructdeepinfra/Qwen/Qwen2.5-VL-32B-Instruct | DeepInfra | $0.20 | $0.60 | 128K | 128K |
| meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8deepinfra/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8 | DeepInfra | $0.20 | $0.80 | 1M | 1M |
| ft:gpt-4.1-nano-2025-04-14ft:gpt-4.1-nano-2025-04-14 | OpenAI | $0.20 | $0.80 | 1M | 32.8K |
| gpt-5.4-nanogpt-5.4-nano | OpenAI | $0.20 | $1.25 | 272K | 128K |
| Qwen/Qwen3-235B-A22B-Instruct-2507-tputtogether_ai/Qwen/Qwen3-235B-A22B-Instruct-2507-tput | together_ai | $0.20 | $6.00 | 262K | — |
| zai-org/GLM-4.5-Air-FP8together_ai/zai-org/GLM-4.5-Air-FP8 | together_ai | $0.20 | $1.10 | 131.1K | — |
| mistralai/Mistral-7B-Instruct-v0.1together_ai/mistralai/Mistral-7B-Instruct-v0.1 | together_ai | $0.20 | $0.20 | 32.8K | — |
| Qwen/Qwen3-235B-A22B-fp8-tputtogether_ai/Qwen/Qwen3-235B-A22B-fp8-tput | together_ai | $0.20 | $0.60 | 40K | — |
| together-ai-4.1b-8btogether-ai-4.1b-8b | together_ai | $0.20 | $0.20 | — | — |
| meta-llama/llama-4-maverick-17b-128e-instructgroq/meta-llama/llama-4-maverick-17b-128e-instruct | Groq | $0.20 | $0.60 | 131.1K | 8.2K |
| meta-llama/llama-guard-4-12bgroq/meta-llama/llama-guard-4-12b | Groq | $0.20 | $0.20 | 8.2K | 8.2K |
| gpt-5.4-nanoazure/gpt-5.4-nano | Azure | $0.20 | $1.25 | 272K | 128K |
| codestral@latestvertex_ai/codestral@latest | $0.20 | $0.60 | 128K | 128K | |
| codestral@2405vertex_ai/codestral@2405 | $0.20 | $0.60 | 128K | 128K | |
| codestral-2501vertex_ai/codestral-2501 | $0.20 | $0.60 | 128K | 128K | |
| Qwen/Qwen3-235B-A22Bnebius/Qwen/Qwen3-235B-A22B | nebius | $0.20 | $0.60 | 262.1K | 262.1K |
| Qwen/QwQ-32Bhyperbolic/Qwen/QwQ-32B | hyperbolic | $0.20 | $0.20 | 131.1K | 131.1K |
| deepseek-ai/DeepSeek-V3hyperbolic/deepseek-ai/DeepSeek-V3 | hyperbolic | $0.20 | $0.20 | 32.8K | 32.8K |
| nvidia.nemotron-nano-12b-v2nvidia.nemotron-nano-12b-v2 | bedrock_converse | $0.20 | $0.60 | 128K | 8K |
| mistral.ministral-3-14b-instructmistral.ministral-3-14b-instruct | bedrock_converse | $0.20 | $0.20 | 128K | 8K |
| moonshot-v1-8k-vision-previewmoonshot/moonshot-v1-8k-vision-preview | moonshot | $0.20 | $2.00 | 8.2K | 8.2K |
| kimi-latest-8kmoonshot/kimi-latest-8k | moonshot | $0.20 | $2.00 | 8.2K | 8.2K |
| moonshot-v1-8kmoonshot/moonshot-v1-8k | moonshot | $0.20 | $2.00 | 8.2K | 8.2K |
| moonshot-v1-8k-0430moonshot/moonshot-v1-8k-0430 | moonshot | $0.20 | $2.00 | 8.2K | 8.2K |
| jamba-mini-1.7jamba-mini-1.7 | AI21 | $0.20 | $0.40 | 256K | 256K |
| jamba-1.5jamba-1.5 | AI21 | $0.20 | $0.40 | 256K | 256K |
| jamba-1.5-minijamba-1.5-mini | AI21 | $0.20 | $0.40 | 256K | 256K |
| jamba-mini-1.6jamba-mini-1.6 | AI21 | $0.20 | $0.40 | 256K | 256K |
| jamba-1.5-mini@001jamba-1.5-mini@001 | AI21 | $0.20 | $0.40 | 256K | 256K |
| jamba-1.5vertex_ai/jamba-1.5 | $0.20 | $0.40 | 256K | 256K | |
| jamba-1.5-mini@001vertex_ai/jamba-1.5-mini@001 | $0.20 | $0.40 | 256K | 256K | |
| jamba-1.5-minivertex_ai/jamba-1.5-mini | $0.20 | $0.40 | 256K | 256K | |
| ai21.jamba-1-5-mini-v1:0ai21.jamba-1-5-mini-v1:0 | Amazon | $0.20 | $0.40 | 256K | 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 |
|---|---|---|
| meta-llama/Llama-3-8b-chat-hf | together_ai | $0.0600 |
| coding-xiaomi-mimo-v2.5-pro | aihubmix | $0.0800 |
| gpt-5.6-luna | aihubmix | $0.1600 |
| gpt-5.4-nano | aihubmix | $0.1650 |
| gpt-5.6-luna-2026-07-09 | Azure | $0.1600 |
| stepfun/step-3.7-flash | openrouter | $0.1550 |
| openai/gpt-4.1-mini:batch | openrouter | $0.1200 |
| mistralai/mistral-medium-3.1:batch | openrouter | $0.1400 |
| cognitivecomputations/dolphin-mistral-24b-venice-edition | openrouter | $0.1300 |
| llama-guard-3-8b | Groq | $0.0600 |
| llama-3.1-8b-instruct | Perplexity | $0.0600 |
| zai/glm-4.5-air | vercel_ai_gateway | $0.1500 |
| alibaba/qwen-3-235b | vercel_ai_gateway | $0.1000 |
| meta/llama-4-maverick | vercel_ai_gateway | $0.1000 |
| google/gemma-2-9b | vercel_ai_gateway | $0.0600 |
| mistralai/ministral-14b-2512 | openrouter | $0.0600 |
| deepseek-ai/DeepSeek-R1-Distill-Llama-70B | DeepInfra | $0.1000 |
| Qwen/Qwen2.5-VL-32B-Instruct | DeepInfra | $0.1000 |
| meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8 | DeepInfra | $0.1200 |
| ft:gpt-4.1-nano-2025-04-14 | OpenAI | $0.1200 |
| gpt-5.4-nano | OpenAI | $0.1650 |
| Qwen/Qwen3-235B-A22B-Instruct-2507-tput | together_ai | $0.6400 |
| zai-org/GLM-4.5-Air-FP8 | together_ai | $0.1500 |
| mistralai/Mistral-7B-Instruct-v0.1 | together_ai | $0.0600 |
| Qwen/Qwen3-235B-A22B-fp8-tput | together_ai | $0.1000 |
| together-ai-4.1b-8b | together_ai | $0.0600 |
| meta-llama/llama-4-maverick-17b-128e-instruct | Groq | $0.1000 |
| meta-llama/llama-guard-4-12b | Groq | $0.0600 |
| gpt-5.4-nano | Azure | $0.1650 |
| codestral@latest | $0.1000 |