Sep 24, 2026 GLM 5.1 Input $0.965 to $1.4, Output $3.03 to $4.4, Cached input $0.179 to $0.26 on baidu/fp8 on baidu/fp8 Input $0.965 Output $3.03 Cached input $0.179 Input $1.4 Output $4.4 Cached input $0.26 +45%Sep 24, 2026 Kimi K2.6 Input $0.432 to $0.95, Output $1.82 to $4, Cached input $0.073 to $0.16 on baidu/fp4 on baidu/fp4 Input $0.432 Output $1.82 Cached input $0.073 Input $0.95 Output $4 Cached input $0.16 +120%Sep 24, 2026 Kimi K2.6 Input $0.432 to $0.497, Output $2.38 to $2.97, Cached input $0.117 to $0.128 on inceptron/int4 on inceptron/int4 Input $0.432 Output $2.38 Cached input $0.117 Input $0.497 Output $2.97 Cached input $0.128 +15%Sep 24, 2026 DeepSeek V4 Flash 0423 Input $0.084 to $0.14, Output $0.168 to $0.28, Cached input $0.017 to $0.028 on baidu/fp8 on baidu/fp8 Input $0.084 Output $0.168 Cached input $0.017 Input $0.14 Output $0.28 Cached input $0.028 +67%Sep 24, 2026 DeepSeek V4 Flash 0423 Input $0.082 to $0.089, Output $0.165 to $0.177, Cached input $0.016 to $0.018 on streamlake/fp8 on streamlake/fp8 Input $0.082 Output $0.165 Cached input $0.016 Input $0.089 Output $0.177 Cached input $0.018 +7%Sep 24, 2026 DeepSeek V4 Pro 0423 Input $0.948 to $1.69, Output $1.9 to $3.38, Cached input $0.079 to $0.14 on baidu/fp8 on baidu/fp8 Input $0.948 Output $1.9 Cached input $0.079 Input $1.69 Output $3.38 Cached input $0.14 +78%Sep 24, 2026 DeepSeek V4 Pro 0423 Input $0.946 to $0.94, Output $1.89 to $1.88, Cached input $0.079 to $0.078 on streamlake/fp8 on streamlake/fp8 Input $0.946 Output $1.89 Cached input $0.079 Input $0.94 Output $1.88 Cached input $0.078 -1%Sep 24, 2026 GLM 5.2 Input $0.648 to $2.25, Output $2.27 to $7.88, Cached input $0.161 to $0.56 on baidu/fast on baidu/fast Input $0.648 Output $2.27 Cached input $0.161 Input $2.25 Output $7.88 Cached input $0.56 +247%Sep 24, 2026 GLM 5.2 Cached input $0.3 to $0.26 on wafer on wafer Cached input $0.3 Cached input $0.26 -13%Sep 24, 2026 GLM 5.2 Input $0.561 to $1.4, Output $1.76 to $4.4, Cached input $0.104 to $0.26 on baidu/fp8 on baidu/fp8 Input $0.561 Output $1.76 Cached input $0.104 Input $1.4 Output $4.4 Cached input $0.26 +149%Sep 24, 2026 GLM 5.2 Input $1.01 to $1.01, Output $3.18 to $3.24, Cached input $0.193 to $0.194 on inceptron/fp4 on inceptron/fp4 Input $1.01 Output $3.18 Cached input $0.193 Input $1.01 Output $3.24 Cached input $0.194 +0%Sep 24, 2026 GLM 5.2 Input $0.641 to $0.643, Output $2.01 to $2.02, Cached input $0.119 to $0.119 on streamlake/fp8 on streamlake/fp8 Input $0.641 Output $2.01 Cached input $0.119 Input $0.643 Output $2.02 Cached input $0.119 +0%Sep 24, 2026 Kimi K3 Input $2.49 to $1.99 on wafer on wafer Input $2.49 Input $1.99 -20%Sep 24, 2026 Kimi K3 Input $1.95 to $0.91, Output $9.75 to $6.99 on inferencenet/fp4 on inferencenet/fp4 Input $1.95 Output $9.75 Input $0.91 Output $6.99 -53%Sep 24, 2026 DeepSeek V4 Flash 0731 Input $0.176 to $0.352, Output $0.528 to $1.06, Cached input $0.018 to $0.035 on qwen on qwen Input $0.176 Output $0.528 Cached input $0.018 Input $0.352 Output $1.06 Cached input $0.035 +100%Sep 24, 2026 DeepSeek V4 Flash 0731 Input $0.087 to $0.08, Cached input $0.08 to $0.07 on wafer/fast on wafer/fast Input $0.087 Cached input $0.08 Input $0.08 Cached input $0.07 -8%Sep 24, 2026 DeepSeek V4 Pro 0813 Input $0.66 to $1.32, Output $1.98 to $3.96, Cached input $0.022 to $0.044 Input $0.66 Output $1.98 Cached input $0.022 Input $1.32 Output $3.96 Cached input $0.044 +100%Sep 24, 2026 DeepSeek V4 Pro 0813 Input $0.4 to $1.32, Output $1.2 to $3.96, Cached input $0.013 to $0.042 on baidu/fp8 on baidu/fp8 Input $0.4 Output $1.2 Cached input $0.013 Input $1.32 Output $3.96 Cached input $0.042 +230%Sep 24, 2026 DeepSeek V4 Pro 0813 Input $0.581 to $1.12, Output $1.74 to $3.37, Cached input $0.058 to $0.112 on qwen on qwen Input $0.581 Output $1.74 Cached input $0.058 Input $1.12 Output $3.37 Cached input $0.112 +93%Sep 24, 2026 DeepSeek V4 Pro 0813 Input $0.395 to $0.387 on wafer on wafer Input $0.395 Input $0.387 -2%Sep 24, 2026 Qwen3.8 27B Input $0.121 to $0.099 on wafer on wafer Input $0.121 Input $0.099 -18%Sep 24, 2026 GLM 5.3 Input $0.561 to $1.4, Output $1.76 to $4.4, Cached input $0.104 to $0.26 on baidu/fp8 on baidu/fp8 Input $0.561 Output $1.76 Cached input $0.104 Input $1.4 Output $4.4 Cached input $0.26 +149%Sep 24, 2026 GLM 5.3 Input $1.19 to $0.99, Cached input $0.36 to $0.26 on wafer on wafer Input $1.19 Cached input $0.36 Input $0.99 Cached input $0.26 -17%Sep 24, 2026 GLM 5.3 Input $0.786 to $0.89, Output $2.65 to $3.73, Cached input $0.203 to $0.257 on inceptron/fp4 on inceptron/fp4 Input $0.786 Output $2.65 Cached input $0.203 Input $0.89 Output $3.73 Cached input $0.257 +13%Sep 24, 2026 GLM 5.3 Flash Input $0.1 to $0.09, Output $0.5 to $0.36, Cached input $0.025 to $0.02 on openinference/fp4 on openinference/fp4 Input $0.1 Output $0.5 Cached input $0.025 Input $0.09 Output $0.36 Cached input $0.02 -10%Sep 24, 2026 DeepSeek V4.1 Flash Input $0.15 to $0.3, Output $0.6 to $1.2, Cached input $0.003 to $0.006 Input $0.15 Output $0.6 Cached input $0.003 Input $0.3 Output $1.2 Cached input $0.006 +100%Sep 24, 2026 DeepSeek V4.1 Flash Input $0.15 to $0.3, Output $0.6 to $1.2, Cached input $0.015 to $0.03 on qwen on qwen Input $0.15 Output $0.6 Cached input $0.015 Input $0.3 Output $1.2 Cached input $0.03 +100%Sep 24, 2026 DeepSeek V4.1 Flash Input $0.285 to $0.225, Output $1.14 to $0.9, Cached input $0.0057 to $0.0045 on gmicloud/fp8 on gmicloud/fp8 Input $0.285 Output $1.14 Cached input $0.0057 Input $0.225 Output $0.9 Cached input $0.0045 -21%Sep 24, 2026 DeepSeek V4.1 Flash Input $0.15 to $0.199, Output $0.6 to $0.794, Cached input $0.003 to $0.004 on streamlake/fp8 on streamlake/fp8 Input $0.15 Output $0.6 Cached input $0.003 Input $0.199 Output $0.794 Cached input $0.004 +32%Sep 24, 2026 DeepSeek V4.1 Flash Input $0.169 to $0.099 on wafer on wafer Input $0.169 Input $0.099 -41%