Fireworks AI logo

Fireworks AI

模型集成商

Fireworks AI is a high-performance serverless inference platform for open and open-weight frontier models (DeepSeek, GLM, Qwen, Kimi, MiniMax, GPT-OSS), offering Standard and Priority throughput tiers with cached-input discounts.

模型
29
提示缓存
批量折扣
Tool Calling
模型29
DeepSeek V4 FlashDeepSeek¥0.9395¥1.88
DeepSeek V4 Flash (0731)DeepSeek¥1.48¥4.43
DeepSeek V4 Flash Vision ExpDeepSeek¥1.48¥4.43
DeepSeek V4 ProDeepSeek¥11.68¥23.35
DeepSeek V4 Pro (0813)DeepSeek¥8.86¥26.58
GLM 5.1GLM¥9.4¥29.53
GLM 5.1 FastGLM¥18.79¥59.06
GLM 5.2GLM¥9.4¥29.53
GLM 5.2 FastGLM¥14.09¥44.29
GLM 5.2 Fast USGLM¥14.09¥44.29
GLM 5.3GLM¥9.4¥29.53
GLM 5.3 FastGLM¥14.09¥44.29
GLM 5.3 FlashGLM¥1.01¥3.36
Kimi K2.6Kimi¥6.38¥26.84
Kimi K2.6 FastKimi¥13.42¥53.69
Kimi K2.7 CodeKimi¥6.38¥26.84
Kimi K2.7 Code FastKimi¥12.75¥53.69
Kimi K3Kimi¥20.13¥100.66
Kimi K3 FastKimi¥30.2¥151
Kimi K3 USKimi¥22.15¥110.73
MiniMax M2.7MiniMax¥2.01¥8.05
MiniMax M3MiniMax¥2.01¥8.05
Muse Glimmer 30B¥2.35¥10.07
NVIDIA Nemotron 3 Ultra (Preview)Nemotron¥4.03¥16.11
NVIDIA Nemotron 3.5 Lightning 30B A3BNemotron¥0.3355¥1.34
OpenAI GPT OSS 120BGPT-OSS¥1.01¥4.03
OpenAI GPT OSS 20BGPT-OSS¥0.4698¥2.01
Qwen 3.7 PlusQwen¥2.68¥10.74
Qwen 3.8 MaxQwen¥13.42¥40.27