大模型 API 价格变动时间线(降价/调价记录)· 大模型库

每日自动比对各模型价格,有变动即记录在此。绿色为降价、红色为涨价; 点击模型名查看该模型完整价格历史曲线。← 返回总表

2026-10-23

Mistral: Mistral Small 3.2 24B输出价$0.25 → $0.20-20%
Mistral: Mistral Small 3.2 24B输入价$0.09 → $0.07-20%
Qwen: Qwen3.5-122B-A10B输出价$2.08 → $3.20+54%
Qwen: Qwen3.5-122B-A10B输入价$0.26 → $0.40+54%
Qwen: Qwen3.5-9B最大输出66K → 236K
Qwen: Qwen3.5-9B输出价$0.13 → $0.25+92%
Qwen: Qwen3.5-9B输入价$0.08 → $0.17+113%
Google: Gemma 4 26B A4B 缓存读$0.04 → $0.04-12%
Google: Gemma 4 26B A4B 输出价$0.26 → $0.23-12%
Google: Gemma 4 26B A4B 输入价$0.08 → $0.07-12%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.04 → $0.06+56%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.89 → $1.39+56%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.44 → $0.69+56%
MoonshotAI: Kimi K2.7 Code缓存读$0.18 → $0.15-17%
MoonshotAI: Kimi K3输出价$11.94 → $10.80-10%
Qwen: Qwen3.8 27B最大输出66K → 236K
Qwen: Qwen3.8 27B输出价$1.35 → $2.50+85%
Z.ai: GLM 5.3最大输出262K → 944K
Z.ai: GLM 5.3缓存读$0.13 → $0.08-40%
Z.ai: GLM 5.3输出价$2.20 → $1.40-36%
Z.ai: GLM 5.3 Flash最大输出944K → 944K
Z.ai: GLM 5.3 Flash输出价$0.20 → $0.50+152%
DeepSeek: DeepSeek V4.1 Flash缓存读$0.00 → $0.01+757%
DeepSeek: DeepSeek V4.1 Flash输出价$0.14 → $1.20+777%
DeepSeek: DeepSeek V4.1 Flash输入价$0.02 → $0.30+1799%
MiniMax: MiniMax M3.1 Flash上下文窗口1M → 1.05M

2026-10-22

Qwen: Qwen3 Next 80B A3B Instruct缓存读$0.07 → —
Qwen: Qwen3.5 397B A17B最大输出66K → 236K
Qwen: Qwen3.5 397B A17B缓存读— → $0.30
Qwen: Qwen3.5 397B A17B输出价$2.34 → $3.60+54%
Qwen: Qwen3.5 397B A17B输入价$0.39 → $0.50+28%
NVIDIA: Nemotron 3 Super输出价$0.45 → $2.40+433%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.03 → $0.04+15%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.77 → $0.89+15%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.38 → $0.44+15%
Z.ai: GLM 5.2输入价$0.62 → $0.03-94%
Z.ai: GLM 5.3输入价$0.08 → $0.70+786%
Z.ai: GLM 5.3 Flash最大输出944K → 944K
Z.ai: GLM 5.3 Flash输出价$0.50 → $0.20-60%
DeepSeek: DeepSeek V4.1 Flash最大输出393K → 944K
TypeSafe: Jev Router最大输出— → 66K
Reflection: Beam 1.0 501B A23B最大输出944K → 131K
Reflection: Beam 1.0 501B A23B缓存读$0.04 → $0.05+25%
Reflection: Beam 1.0 501B A23B输入价$0.30 → $0.25-17%

2026-10-21

Qwen: Qwen3 Next 80B A3B Instruct最大输出236K → 16K
Qwen: Qwen3 Next 80B A3B Instruct输入价$0.10 → $0.09-10%
Qwen: Qwen3.5 397B A17B缓存读$0.30 → —
Qwen: Qwen3.5-9B最大输出33K → 66K
Qwen: Qwen3.5-9B缓存读— → $0.04
Qwen: Qwen3.5-9B输出价$0.15 → $0.13-13%
Qwen: Qwen3.5-9B输入价$0.10 → $0.08-20%
NVIDIA: Nemotron 3 Super缓存读— → $0.15
NVIDIA: Nemotron 3 Super输入价$0.08 → $0.30+275%
Z.ai: GLM 5.1最大输出128K → 131K
Z.ai: GLM 5.1缓存读$0.18 → $0.26+45%
Z.ai: GLM 5.1输出价$3.04 → $4.40+45%
Z.ai: GLM 5.1输入价$0.97 → $1.40+45%
Xiaomi: MiMo-V2.5最大输出131K → 945K
Xiaomi: MiMo-V2.5缓存读$0.00 → $0.00-7%
Xiaomi: MiMo-V2.5输出价$0.28 → $0.24-15%
Xiaomi: MiMo-V2.5输入价$0.14 → $0.12-15%
Xiaomi: MiMo-V2.5-Pro最大输出131K → 945K
Xiaomi: MiMo-V2.5-Pro缓存读$0.00 → $0.00-22%
Xiaomi: MiMo-V2.5-Pro输出价$0.87 → $0.61-30%
Xiaomi: MiMo-V2.5-Pro输入价$0.43 → $0.30-30%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.03 → $0.03+14%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.67 → $0.77+14%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.34 → $0.38+14%
Z.ai: GLM 5.2缓存读$0.14 → $0.03-76%
Z.ai: GLM 5.2输出价$2.42 → $7.00+189%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.00 → $0.00-66%
DeepSeek: DeepSeek V4 Flash 0731输出价$0.39 → $0.13-66%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.13 → $0.04-66%
DeepSeek: DeepSeek V4 Pro 0813最大输出384K → 393K
Qwen: Qwen3.8 2.4T A95B最大输出131K → 909K
Qwen: Qwen3.8 27B输入价$0.24 → $0.04-83%
Z.ai: GLM 5.3最大输出944K → 262K
Z.ai: GLM 5.3缓存读$0.08 → $0.13+67%
Z.ai: GLM 5.3输出价$1.40 → $2.20+57%
Z.ai: GLM 5.3 Flash最大输出944K → 944K
Z.ai: GLM 5.3 Flash输出价$0.12 → $0.50+317%
Z.ai: GLM 5.3 Flash输入价$0.05 → $0.15+218%
DeepSeek: DeepSeek V4.1 Flash最大输出944K → 393K
DeepSeek: DeepSeek V4.1 Flash缓存读$0.00 → $0.00+17%
DeepSeek: DeepSeek V4.1 Flash输出价$0.12 → $0.14+14%

2026-10-20

DeepSeek: DeepSeek V3最大输出16K → 16K
DeepSeek: DeepSeek V3输出价$0.89 → $1.03+16%
DeepSeek: DeepSeek V3输入价$0.32 → $0.26-20%
Google: Gemma 3 4B最大输出16K → 118K
Google: Gemma 3 4B缓存读— → $0.02
Google: Gemma 3 4B输出价$0.10 → $0.05-50%
Google: Gemma 3 4B输入价$0.05 → $0.04-20%
Qwen: Qwen3 Next 80B A3B Instruct最大输出16K → 236K
Qwen: Qwen3 Next 80B A3B Instruct缓存读— → $0.07
Qwen: Qwen3 Next 80B A3B Instruct输入价$0.09 → $0.10+11%
Qwen: Qwen3.5 397B A17B最大输出236K → 66K
Qwen: Qwen3.5 397B A17B输出价$3.60 → $2.34-35%
Qwen: Qwen3.5 397B A17B输入价$0.50 → $0.39-22%
Qwen: Qwen3.5-35B-A3B最大输出236K → 82K
Qwen: Qwen3.5-35B-A3B输入价$0.15 → $0.14-7%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.03 → $0.03-8%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.73 → $0.67-7%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.36 → $0.34-7%
Qwen: Qwen3.6 35B A3B输出价$1.00 → $0.95-5%
Qwen: Qwen3.6 35B A3B输入价$0.15 → $0.10-33%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.01 → $0.13+1409%
Upstage: Solar Pro 4缓存读$0.02 → $0.03+67%
Upstage: Solar Pro 4输出价$0.36 → $0.60+67%
Upstage: Solar Pro 4输入价$0.09 → $0.15+67%
Qwen: Qwen3.8 27B最大输出236K → 66K
Qwen: Qwen3.8 27B缓存读$0.05 → $0.02-64%
Qwen: Qwen3.8 27B输出价$2.20 → $1.35-39%
Z.ai: GLM 5.3 Flash输出价$0.16 → $0.12-25%
Z.ai: GLM 5.3 Flash输入价$0.05 → $0.05-2%
DeepSeek: DeepSeek V4.1 Flash缓存读$0.00 → $0.00-62%
Mistral: Mistral Large 4缓存读$0.07 → $0.14+100%
Mistral: Mistral Large 4输出价$2.09 → $4.18+100%
Mistral: Mistral Large 4输入价$0.68 → $1.36+100%
NVIDIA: Nemotron 3.5 Super VL缓存读$0.08 → $0.02-75%
NVIDIA: Nemotron 3.5 Super VL输出价$0.40 → $0.50+25%
NVIDIA: Nemotron 3.5 Super VL输入价$0.10 → $0.15+50%

2026-10-19

Meta: Llama 4 Maverick上下文窗口1.05M → 524K
Z.ai: GLM 4.6缓存读$0.10 → $0.11+10%
Z.ai: GLM 4.6输出价$2.00 → $2.20+10%
Z.ai: GLM 4.6输入价$0.50 → $0.55+10%
Qwen: Qwen3.5-27B最大输出82K → 66K
Qwen: Qwen3.5-27B缓存读— → $0.27
Qwen: Qwen3.5-27B输出价$2.60 → $2.16-17%
Qwen: Qwen3.5-27B输入价$0.26 → $0.27+4%
Google: Gemma 4 26B A4B 缓存读$0.04 → $0.04+13%
Google: Gemma 4 26B A4B 输出价$0.23 → $0.26+13%
Google: Gemma 4 26B A4B 输入价$0.07 → $0.08+13%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.02 → $0.03+74%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.42 → $0.73+74%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.21 → $0.36+74%
MoonshotAI: Kimi K3输入价$0.25 → $0.24-4%
DeepSeek: DeepSeek V4 Flash 0731最大输出944K → 131K
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.01 → $0.00-52%
DeepSeek: DeepSeek V4 Flash 0731输出价$0.45 → $0.39-13%
Z.ai: GLM 5.3 Flash输出价$0.20 → $0.16-20%
Google: Gemini 4 Argon最大输出66K → 944K
Reflection: Beam 1.0 501B A23B缓存读$0.10 → $0.04-60%
Reflection: Beam 1.0 501B A23B输出价$3.00 → $1.00-67%
Reflection: Beam 1.0 501B A23B输入价$1.00 → $0.30-70%

2026-10-18

Meta: Llama 3.2 3B Instruct最大输出118K → 72K
Meta: Llama 3.2 3B Instruct输出价$0.33 → $0.34+2%
Meta: Llama 3.2 3B Instruct输入价$0.05 → $0.05+2%
Google: Gemma 3 27B输出价$0.45 → $0.35-22%
Morph: Morph V3 Fast最大输出38K → 131K
Morph: Morph V3 Fast上下文窗口82K → 262K
Qwen: Qwen3 30B A3B Instruct 2507输入价$0.10 → $0.09-10%
DeepSeek: DeepSeek V3.2缓存读$0.14 → $0.03-80%
MoonshotAI: Kimi K3输入价$0.30 → $0.25-18%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.01 → $0.01-14%
DeepSeek: DeepSeek V4 Flash 0731输出价$1.28 → $0.45-65%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.01 → $0.01-14%
DeepSeek: DeepSeek V4 Pro 0813最大输出393K → 384K
Z.ai: GLM 5.3输出价$4.80 → $1.40-71%
Z.ai: GLM 5.3 Flash输入价$0.15 → $0.05-68%
DeepSeek: DeepSeek V4.1 Flash输出价$0.14 → $0.12-17%
NVIDIA: Nemotron 3.5 Super VL缓存读$0.02 → $0.08+300%
NVIDIA: Nemotron 3.5 Super VL输出价$0.50 → $0.40-20%
NVIDIA: Nemotron 3.5 Super VL输入价$0.15 → $0.10-33%

2026-10-17

Meta: Llama 3.3 70B Instruct缓存读$0.29 → —
Meta: Llama 3.3 70B Instruct输入价$0.59 → $0.10-83%
Qwen: Qwen3 Next 80B A3B Instruct缓存读$0.07 → —
DeepSeek: DeepSeek V3.2最大输出147K → 66K
DeepSeek: DeepSeek V3.2输出价$0.80 → $0.42-47%
DeepSeek: DeepSeek V3.2输入价$0.26 → $0.28+8%
Z.ai: GLM 4.7缓存读$0.11 → $0.10-10%
Z.ai: GLM 4.7输出价$2.20 → $1.98-10%
Z.ai: GLM 4.7输入价$0.60 → $0.54-10%
Qwen: Qwen3.5-35B-A3B最大输出16K → 236K
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.02 → $0.02-25%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.55 → $0.42-25%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.28 → $0.21-25%
Z.ai: GLM 5.2输入价$0.04 → $0.62+1576%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.02 → $0.01-35%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.02 → $0.01-35%
Z.ai: GLM 5.3缓存读$0.40 → $0.08-80%
Z.ai: GLM 5.3输入价$0.40 → $0.08-80%
Z.ai: GLM 5.3 Flash最大输出131K → 944K
Z.ai: GLM 5.3 Flash缓存读$0.01 → $0.01-52%
Z.ai: GLM 5.3 Flash输出价$0.71 → $0.20-72%
DeepSeek: DeepSeek V4.1 Flash缓存读$0.01 → $0.00-70%
DeepSeek: DeepSeek V4.1 Flash输出价$0.45 → $0.14-68%

2026-10-16

Meta: Llama 3.3 70B Instruct最大输出33K → 16K
Meta: Llama 3.3 70B Instruct输出价$0.79 → $0.32-59%
Meta: Llama 4 Scout最大输出118K → 16K
Meta: Llama 4 Scout输出价$0.59 → $0.30-49%
Meta: Llama 4 Scout输入价$0.18 → $0.10-44%
Qwen: Qwen3 Next 80B A3B Instruct最大输出236K → 16K
Qwen: Qwen3 Next 80B A3B Instruct输入价$0.10 → $0.09-10%
Google: Gemma 4 26B A4B 缓存读$0.04 → $0.04-12%
Google: Gemma 4 26B A4B 输出价$0.26 → $0.23-12%
Google: Gemma 4 26B A4B 输入价$0.08 → $0.07-12%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.03 → $0.02-21%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.70 → $0.55-21%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.35 → $0.28-21%
Z.ai: GLM 5.2缓存读$0.04 → $0.14+289%
Z.ai: GLM 5.2输出价$7.00 → $2.42-65%
MoonshotAI: Kimi K3缓存读$0.28 → $0.21-25%
MoonshotAI: Kimi K3输出价$15.00 → $11.94-20%
MoonshotAI: Kimi K3输入价$0.41 → $0.30-25%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.01 → $0.02+18%
DeepSeek: DeepSeek V4 Flash 0731输出价$0.64 → $1.28+100%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.01 → $0.02+18%
Qwen: Qwen3.8 27B输入价$0.04 → $0.24+500%
Z.ai: GLM 5.3输出价$8.00 → $4.80-40%
Z.ai: GLM 5.3 Flash最大输出944K → 131K
Z.ai: GLM 5.3 Flash缓存读$0.03 → $0.01-58%
Z.ai: GLM 5.3 Flash输出价$0.50 → $0.71+42%
DeepSeek: DeepSeek V4.1 Flash最大输出393K → 944K
DeepSeek: DeepSeek V4.1 Flash缓存读$0.00 → $0.01+80%
DeepSeek: DeepSeek V4.1 Flash输出价$0.60 → $0.45-25%

2026-10-15

Mistral: Mistral Small 3.2 24B输出价$0.20 → $0.25+25%
Mistral: Mistral Small 3.2 24B输入价$0.07 → $0.09+25%
OpenAI: gpt-oss-120b最大输出118K → 66K
OpenAI: gpt-oss-120b输出价$0.17 → $0.60+253%
OpenAI: gpt-oss-120b输入价$0.04 → $0.15+305%
Z.ai: GLM 5最大输出128K → 131K
Z.ai: GLM 5缓存读$0.12 → $0.20+67%
Z.ai: GLM 5输出价$1.92 → $3.20+67%
Z.ai: GLM 5输入价$0.60 → $1.00+67%
MiniMax: MiniMax M2.5最大输出128K → 33K
MiniMax: MiniMax M2.5缓存读$0.03 → $0.03+11%
MiniMax: MiniMax M2.5输出价$1.08 → $0.95-12%
Google: Gemma 4 31B缓存读$0.05 → —
Z.ai: GLM 5.1最大输出131K → 128K
Z.ai: GLM 5.1缓存读$0.25 → $0.18-27%
Z.ai: GLM 5.1输出价$4.18 → $3.04-27%
Z.ai: GLM 5.1输入价$1.33 → $0.97-27%
Qwen: Qwen3.6 27B最大输出82K → 66K
Qwen: Qwen3.6 27B缓存读— → $0.03
Qwen: Qwen3.6 27B输出价$3.20 → $2.00-38%
Qwen: Qwen3.6 27B输入价$0.32 → $0.25-22%
Z.ai: GLM 5.2缓存读$0.04 → $0.04-8%
Z.ai: GLM 5.2输入价$0.04 → $0.04-8%
MoonshotAI: Kimi K3输入价$0.58 → $0.41-30%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.02 → $0.01-18%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.02 → $0.01-18%
Qwen: Qwen3.8 27B缓存读$0.04 → $0.05+28%
Qwen: Qwen3.8 27B输出价$2.30 → $2.20-4%
Z.ai: GLM 5.3缓存读$0.12 → $0.40+235%
Z.ai: GLM 5.3输出价$4.80 → $8.00+67%
Z.ai: GLM 5.3输入价$0.12 → $0.40+233%
Google: Gemini 3.8 Flash缓存读$0.07 → $0.08+10%
Google: Gemini 3.8 Flash输出价$3.75 → $4.13+10%
Google: Gemini 3.8 Flash输入价$0.75 → $0.82+10%
DeepSeek: DeepSeek V4.1 Flash最大输出944K → 393K
DeepSeek: DeepSeek V4.1 Flash缓存读$0.01 → $0.00-44%
DeepSeek: DeepSeek V4.1 Flash输出价$0.89 → $0.60-33%
StepFun: Step 5 Preview上下文窗口1M → 1.05M
StepFun: Step 5 Preview缓存读$0.05 → $0.10+100%

2026-10-14

Mistral: Mistral Small 3.2 24B输出价$0.25 → $0.20-20%
Mistral: Mistral Small 3.2 24B输入价$0.09 → $0.07-20%
OpenAI: gpt-oss-120b最大输出66K → 118K
OpenAI: gpt-oss-120b输出价$0.60 → $0.17-72%
OpenAI: gpt-oss-120b输入价$0.15 → $0.04-75%
MiniMax: MiniMax M2最大输出177K → 131K
MiniMax: MiniMax M2输出价$1.20 → $1.02-15%
MiniMax: MiniMax M2输入价$0.30 → $0.26-15%
Qwen: Qwen3.5 397B A17B缓存读$0.23 → $0.30+33%
Qwen: Qwen3.5 397B A17B输出价$3.50 → $3.60+3%
Qwen: Qwen3.5 397B A17B输入价$0.55 → $0.50-9%
Google: Gemma 4 31B输出价$0.34 → $0.40+18%
Google: Gemma 4 31B输入价$0.09 → $0.20+122%
Z.ai: GLM 5.1最大输出128K → 131K
Z.ai: GLM 5.1缓存读$0.18 → $0.25+38%
Z.ai: GLM 5.1输出价$3.04 → $4.18+38%
Z.ai: GLM 5.1输入价$0.97 → $1.33+38%
DeepSeek: DeepSeek V4 Flash 0423输出价$1.28 → $0.18-86%
DeepSeek: DeepSeek V4 Flash 0423输入价$0.01 → $0.09+595%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.03 → $0.03-9%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.77 → $0.70-9%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.39 → $0.35-9%
Qwen: Qwen3.6 35B A3B最大输出66K → 236K
Qwen: Qwen3.6 35B A3B缓存读$0.10 → $0.05-50%
Qwen: Qwen3.6 35B A3B输出价$0.95 → $1.00+5%
Qwen: Qwen3.6 35B A3B输入价$0.10 → $0.15+50%
Z.ai: GLM 5.2缓存读$0.07 → $0.04-44%
Z.ai: GLM 5.2输入价$0.08 → $0.04-50%
OpenAI: GPT-5.6 Sol缓存写$2.50 → $5.00+100%
OpenAI: GPT-5.6 Sol缓存读$0.20 → $0.40+100%
OpenAI: GPT-5.6 Sol输出价$10.00 → $20.00+100%
OpenAI: GPT-5.6 Sol输入价$2.00 → $4.00+100%
OpenAI: GPT-5.6 Sol Pro缓存写$2.50 → $5.00+100%
OpenAI: GPT-5.6 Sol Pro缓存读$0.20 → $0.40+100%
OpenAI: GPT-5.6 Sol Pro输出价$10.00 → $20.00+100%
OpenAI: GPT-5.6 Sol Pro输入价$2.00 → $4.00+100%
MoonshotAI: Kimi K3缓存读$0.55 → $0.28-49%
MoonshotAI: Kimi K3输出价$10.00 → $15.00+50%
MoonshotAI: Kimi K3输入价$0.73 → $0.58-21%
Qwen: Qwen3.8 27B缓存读$0.02 → $0.04+97%
Z.ai: GLM 5.3缓存读$0.06 → $0.12+102%
Z.ai: GLM 5.3输入价$0.08 → $0.12+52%
Z.ai: GLM 5.3 Flash最大输出944K → 944K
Z.ai: GLM 5.3 Flash缓存读$0.01 → $0.03+178%
Z.ai: GLM 5.3 Flash输出价$0.90 → $0.50-44%
DeepSeek: DeepSeek V4.1 Flash输出价$0.77 → $0.89+16%
DeepSeek: DeepSeek V4.1 Flash输入价$0.04 → $0.02-64%

2026-10-13

Meta: Llama 3.3 70B Instruct最大输出16K → 33K
Meta: Llama 3.3 70B Instruct缓存读$0.11 → $0.29+168%
Meta: Llama 3.3 70B Instruct输出价$0.50 → $0.79+58%
Meta: Llama 3.3 70B Instruct输入价$0.22 → $0.59+168%
OpenAI: gpt-oss-120b最大输出118K → 66K
OpenAI: gpt-oss-120b缓存读— → $0.07
OpenAI: gpt-oss-120b输出价$0.17 → $0.60+253%
OpenAI: gpt-oss-120b输入价$0.04 → $0.15+305%
Qwen: Qwen3.5-35B-A3B最大输出33K → 16K
Qwen: Qwen3.5-35B-A3B缓存读$0.04 → $0.05+25%
Qwen: Qwen3.5-35B-A3B输出价$0.75 → $1.00+33%
Qwen: Qwen3.5-35B-A3B输入价$0.08 → $0.15+87%
MiniMax: MiniMax M2.7最大输出177K → 131K
MiniMax: MiniMax M2.7缓存读$0.04 → $0.06+43%
MiniMax: MiniMax M2.7输出价$0.84 → $1.20+43%
MiniMax: MiniMax M2.7输入价$0.21 → $0.30+43%
DeepSeek: DeepSeek V4 Flash 0423最大输出944K → 944K
DeepSeek: DeepSeek V4 Flash 0423缓存读$0.01 → $0.02+39%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.03 → $0.03-6%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.82 → $0.77-6%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.41 → $0.39-6%
Z.ai: GLM 5.2缓存读$0.06 → $0.07+19%
Z.ai: GLM 5.2输入价$0.06 → $0.08+33%
Z.ai: GLM 5.3 Flash缓存读$0.02 → $0.01-53%
Z.ai: GLM 5.3 Flash输出价$1.20 → $0.90-25%
DeepSeek: DeepSeek V4.1 Flash最大输出393K → 944K
DeepSeek: DeepSeek V4.1 Flash缓存读$0.00 → $0.01+80%
DeepSeek: DeepSeek V4.1 Flash输出价$0.60 → $0.77+28%
DeepSeek: DeepSeek V4.1 Flash输入价$0.15 → $0.04-71%
inclusionAI: Ling 3.0 Flash VL上下文窗口262K → 131K
inclusionAI: Ling 3.0 Flash VL缓存读$0.00 → $0.01+186%
inclusionAI: Ling 3.0 Flash VL输出价$0.06 → $0.18+192%
inclusionAI: Ling 3.0 Flash VL输入价$0.02 → $0.06+186%
inclusionAI: Ling 3.1 Flash缓存读— → $0.01
inclusionAI: Ling 3.1 Flash输出价$0.00 → $0.23
inclusionAI: Ling 3.1 Flash输入价$0.00 → $0.07

2026-10-12

Meta: Llama 4 Scout最大输出16K → 118K
Meta: Llama 4 Scout输出价$0.30 → $0.59+97%
Meta: Llama 4 Scout输入价$0.10 → $0.18+80%
Google: Gemma 4 26B A4B 缓存读$0.04 → $0.04+13%
Google: Gemma 4 26B A4B 输出价$0.23 → $0.26+13%
Google: Gemma 4 26B A4B 输入价$0.07 → $0.08+13%
DeepSeek: DeepSeek V4 Flash 0423缓存读$0.03 → $0.01-56%
DeepSeek: DeepSeek V4 Flash 0423输入价$0.03 → $0.01-56%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.02 → $0.03+98%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.42 → $0.82+98%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.21 → $0.41+98%
MoonshotAI: Kimi K3输出价$13.50 → $10.00-26%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.02 → $0.02-1%
DeepSeek: DeepSeek V4 Flash 0731输出价$0.90 → $0.64-29%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.02 → $0.02-1%
Qwen: Qwen3.8 27B缓存写$0.53 → —
Qwen: Qwen3.8 27B输入价$0.42 → $0.04-91%
Z.ai: GLM 5.3 Flash最大输出944K → 944K
Z.ai: GLM 5.3 Flash缓存读$0.03 → $0.02-23%
Z.ai: GLM 5.3 Flash输出价$0.50 → $1.20+140%
DeepSeek: DeepSeek V4.1 Flash最大输出944K → 393K
DeepSeek: DeepSeek V4.1 Flash缓存读$0.01 → $0.00-50%
DeepSeek: DeepSeek V4.1 Flash输出价$1.20 → $0.60-50%
DeepSeek: DeepSeek V4.1 Flash输入价$0.30 → $0.15-50%
inclusionAI: Ling 3.1 Flash最大输出33K → 236K

2026-10-11

MoonshotAI: Kimi K2.6缓存读$0.11 → $0.15+39%
MoonshotAI: Kimi K2.6输出价$2.45 → $3.41+39%
MoonshotAI: Kimi K2.6输入价$0.43 → $0.65+50%
NVIDIA: Nemotron 3 Ultra最大输出16K → 183K
NVIDIA: Nemotron 3 Ultra缓存读$0.10 → $0.12+20%
NVIDIA: Nemotron 3 Ultra输出价$2.20 → $2.40+9%
NVIDIA: Nemotron 3 Ultra输入价$0.50 → $0.60+20%
MoonshotAI: Kimi K3输出价$15.00 → $13.50-10%
MoonshotAI: Kimi K3输入价$0.80 → $0.73-9%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.01 → $0.02+69%
DeepSeek: DeepSeek V4 Flash 0731输出价$1.28 → $0.90-30%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.01 → $0.02+69%
Qwen: Qwen3.8 27B最大输出131K → 236K
Qwen: Qwen3.8 27B缓存读$0.09 → $0.02-77%
Qwen: Qwen3.8 27B输出价$2.55 → $2.30-10%
Z.ai: GLM 5.3缓存读$0.04 → $0.06+55%
Z.ai: GLM 5.3输入价$0.04 → $0.08+103%

2026-10-10

DeepSeek: DeepSeek V3.2缓存读$0.14 → $0.14+4%
DeepSeek: DeepSeek V3.2输出价$0.42 → $0.80+90%
MoonshotAI: Kimi K2.5缓存读$0.07 → $0.15+114%
MoonshotAI: Kimi K2.5输出价$2.25 → $2.50+11%
MoonshotAI: Kimi K2.5输入价$0.45 → $0.50+11%
DeepSeek: DeepSeek V4 Flash 0423缓存读$0.01 → $0.03+300%
DeepSeek: DeepSeek V4 Flash 0423输入价$0.01 → $0.03+300%
Qwen: Qwen3.6 27B缓存读$0.03 → —
Qwen: Qwen3.6 35B A3B最大输出236K → 66K
Qwen: Qwen3.6 35B A3B缓存读$0.05 → $0.10+100%
Qwen: Qwen3.6 35B A3B输出价$1.00 → $0.95-5%
Qwen: Qwen3.6 35B A3B输入价$0.15 → $0.10-33%
MoonshotAI: Kimi K3缓存读$0.50 → $0.55+10%
MoonshotAI: Kimi K3输出价$13.50 → $15.00+11%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.01 → $0.01-31%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.01 → $0.01-31%
Z.ai: GLM 5.3输出价$6.00 → $4.80-20%
Qwen: Qwen3.8 Flash缓存写$0.20 → —
Qwen: Qwen3.8 Max (0902)缓存写$2.50 → —

2026-10-09

Meta: Llama 3.3 70B Instruct缓存读— → $0.11
Meta: Llama 3.3 70B Instruct输出价$0.32 → $0.50+56%
Meta: Llama 3.3 70B Instruct输入价$0.10 → $0.22+120%
DeepSeek: DeepSeek V3最大输出16K → 16K
DeepSeek: DeepSeek V3输出价$1.03 → $0.89-13%
DeepSeek: DeepSeek V3输入价$0.26 → $0.32+24%
Qwen: Qwen3 14B上下文窗口131K → 41K
Qwen: Qwen3 30B A3B上下文窗口131K → 41K
Qwen: Qwen3 235B A22B Thinking 2507最大输出118K → 16K
Qwen: Qwen3 235B A22B Thinking 2507上下文窗口131K → 128K
Qwen: Qwen3 235B A22B Thinking 2507输出价$2.30 → $3.50+52%
Qwen: Qwen3 235B A22B Thinking 2507输入价$0.23 → $0.45+96%
Qwen: Qwen3 Next 80B A3B Instruct最大输出16K → 236K
Qwen: Qwen3 Next 80B A3B Instruct缓存读— → $0.07
Qwen: Qwen3 Next 80B A3B Instruct输入价$0.09 → $0.10+11%
Qwen: Qwen3 Next 80B A3B Thinking最大输出33K → 236K
Z.ai: GLM 4.6最大输出16K → 131K
Z.ai: GLM 4.6缓存读$0.08 → $0.10+25%
Z.ai: GLM 4.6输出价$1.75 → $2.00+14%
Z.ai: GLM 4.6输入价$0.43 → $0.50+16%
Qwen: Qwen3 VL 30B A3B Thinking最大输出33K → 236K
Qwen: Qwen3 VL 30B A3B Thinking输出价$2.40 → $1.00-58%
Qwen: Qwen3 VL 30B A3B Thinking输入价$0.20 → $0.29+45%
Qwen: Qwen3 VL 8B Instruct最大输出33K → 236K
Qwen: Qwen3 VL 8B Instruct缓存读— → $0.12
Qwen: Qwen3 VL 8B Instruct输出价$0.46 → $0.75+65%
Qwen: Qwen3 VL 8B Instruct输入价$0.12 → $0.25+114%
MoonshotAI: Kimi K2 Thinking最大输出236K → 98K
MoonshotAI: Kimi K2 Thinking缓存读— → $0.15
Z.ai: GLM 4.6V缓存读$0.05 → $0.06+10%
Qwen: Qwen3.5-35B-A3B最大输出236K → 33K
Qwen: Qwen3.5-35B-A3B缓存读$0.05 → $0.04-20%
Qwen: Qwen3.5-35B-A3B输出价$1.00 → $0.75-25%
Qwen: Qwen3.5-35B-A3B输入价$0.15 → $0.08-47%
Google: Gemma 4 26B A4B 缓存读$0.04 → $0.04-12%
Google: Gemma 4 26B A4B 输出价$0.26 → $0.23-12%
Google: Gemma 4 26B A4B 输入价$0.08 → $0.07-12%
MoonshotAI: Kimi K2.6缓存读$0.12 → $0.11-10%
DeepSeek: DeepSeek V4 Flash 0423缓存读$0.01 → $0.01+32%
DeepSeek: DeepSeek V4 Flash 0423输入价$0.01 → $0.01+32%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.02 → $0.02-27%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.57 → $0.42-27%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.29 → $0.21-27%
Qwen: Qwen3.6 27B最大输出66K → 82K
Qwen: Qwen3.6 27B输出价$2.00 → $3.20+60%
Qwen: Qwen3.6 27B输入价$0.30 → $0.32+7%
Z.ai: GLM 5.2缓存读$0.08 → $0.06-29%
Z.ai: GLM 5.2输出价$3.60 → $7.00+94%
Z.ai: GLM 5.2输入价$0.08 → $0.06-29%
MoonshotAI: Kimi K3缓存读$0.55 → $0.50-9%
MoonshotAI: Kimi K3输入价$0.82 → $0.80-2%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.02 → $0.01-24%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.02 → $0.01-24%
NVIDIA: Nemotron 3.5 Lightning最大输出131K → 236K
NVIDIA: Nemotron 3.5 Lightning缓存读$0.02 → $0.04+63%
NVIDIA: Nemotron 3.5 Lightning输出价$0.14 → $0.20+43%
NVIDIA: Nemotron 3.5 Lightning输入价$0.05 → $0.07+43%
Z.ai: GLM 5.3缓存读$0.05 → $0.04-21%
Z.ai: GLM 5.3输出价$3.39 → $6.00+77%
Z.ai: GLM 5.3输入价$0.05 → $0.04-20%

2026-10-08

Meta: Llama 3.3 70B Instruct缓存读$0.11 → —
Qwen: Qwen3 Next 80B A3B Instruct缓存读$0.07 → —
DeepSeek: DeepSeek V4 Flash 0423缓存读$0.01 → $0.01-24%
DeepSeek: DeepSeek V4 Flash 0423输入价$0.01 → $0.01-24%
MoonshotAI: Kimi K3输入价$3.00 → $0.82-73%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.00 → $0.02+291%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.00 → $0.02+291%
Mistral: Mistral Nemo输入价$0.02 → $0.03+53%
Meta: Llama 3.3 70B Instruct输出价$0.50 → $0.32-36%
Meta: Llama 3.3 70B Instruct输入价$0.22 → $0.10-55%
Qwen: Qwen3 30B A3B Instruct 2507最大输出32K → 236K
Qwen: Qwen3 30B A3B Instruct 2507输出价$0.19 → $0.30+55%
Qwen: Qwen3 30B A3B Instruct 2507输入价$0.05 → $0.10+108%
Qwen: Qwen3 Next 80B A3B Instruct最大输出236K → 16K
Qwen: Qwen3 Next 80B A3B Instruct输入价$0.10 → $0.09-10%
NVIDIA: Nemotron 3 Nano 30B A3B最大输出33K → 236K
NVIDIA: Nemotron 3 Nano 30B A3B缓存读$0.03 → —
NVIDIA: Nemotron 3 Nano 30B A3B输出价$0.20 → $0.24+20%
NVIDIA: Nemotron 3 Nano 30B A3B输入价$0.05 → $0.06+20%
Qwen: Qwen3.5 397B A17B最大输出82K → 236K
Qwen: Qwen3.5 397B A17B缓存读$0.22 → $0.23+2%
Qwen: Qwen3.5 397B A17B输出价$3.00 → $3.50+17%
Qwen: Qwen3.5 397B A17B输入价$0.45 → $0.55+22%
Qwen: Qwen3.5-27B最大输出66K → 82K
Qwen: Qwen3.5-27B输出价$1.56 → $2.60+67%
Qwen: Qwen3.5-27B输入价$0.20 → $0.26+33%
MoonshotAI: Kimi K2.6缓存读$0.10 → $0.12+23%
MoonshotAI: Kimi K2.6输入价$0.47 → $0.43-7%
DeepSeek: DeepSeek V4 Flash 0423缓存读$0.03 → $0.01-75%
DeepSeek: DeepSeek V4 Flash 0423输入价$0.03 → $0.01-75%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)缓存读$0.02 → $0.02+37%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输出价$0.42 → $0.57+38%
DeepSeek V4 Pro 0423(旧检查点·第三方美元报价)输入价$0.21 → $0.29+38%
Z.ai: GLM 5.2最大输出131K → 944K
Z.ai: GLM 5.2缓存读$0.08 → $0.08+8%
Z.ai: GLM 5.2输出价$7.20 → $3.60-50%
Z.ai: GLM 5.2输入价$0.09 → $0.08-2%
MoonshotAI: Kimi K3缓存读$0.45 → $0.55+22%
MoonshotAI: Kimi K3输出价$15.00 → $13.50-10%
DeepSeek: DeepSeek V4 Flash 0731缓存读$0.02 → $0.00-74%
DeepSeek: DeepSeek V4 Flash 0731输入价$0.02 → $0.00-74%
Z.ai: GLM 5.3缓存读$0.06 → $0.05-20%

价格数据参考 OpenRouter 及各厂商公开页面整理,以美元 / 百万 tokens 计,可能滞后或有误,请以官方页面为准。不构成任何采购或投资建议。