Fireworks AI models, one API key

24 models served by Fireworks AI are available through FastInfra's OpenAI-compatible endpoint — 14 of them route here by default because it is the cheapest option. Pin it explicitly with the :fireworks suffix.

Read the docs
Glm 5p2 accounts/fireworks/models/glm-5p2 input · output / 1M tokens Kimi K2p6 accounts/fireworks/models/kimi-k2p6 input · output / 1M tokens Kimi K2p7 Code accounts/fireworks/models/kimi-k2p7-code input · output / 1M tokens Minimax M2p7 accounts/fireworks/models/minimax-m2p7 input · output / 1M tokens Nemotron 3 Ultra Nvfp4 accounts/fireworks/models/nemotron-3-ultra-nvfp4 input · output / 1M tokens Nemotron Lightning 3p5 30b a3b accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b input · output / 1M tokens Qwen3p7 Plus accounts/fireworks/models/qwen3p7-plus input · output / 1M tokens Qwen3p8 2p4t A95b accounts/fireworks/models/qwen3p8-2p4t-a95b input · output / 1M tokens Qwen3p8 Max accounts/fireworks/models/qwen3p8-max input · output / 1M tokens Glm 5p2 Fast accounts/fireworks/routers/glm-5p2-fast input · output / 1M tokens Kimi K2p6 Turbo accounts/fireworks/routers/kimi-k2p6-turbo input · output / 1M tokens Kimi K2p7 Code Fast accounts/fireworks/routers/kimi-k2p7-code-fast input · output / 1M tokens Kimi k3 Fast accounts/fireworks/routers/kimi-k3-fast input · output / 1M tokens Deepseek v4 Flash deepseek/deepseek-v4-flash $0.09 input · $0.19 output / 1M tokens Deepseek v4 Flash 0731 deepseek/deepseek-v4-flash-0731 $0.08 input · $0.19 output / 1M tokens Deepseek v4 Pro deepseek/deepseek-v4-pro $1.23 input · $2.45 output / 1M tokens Gpt Oss 120b gpt-oss-120b $0.03 input · $0.18 output / 1M tokens Muse Glimmer 30b meta/muse-glimmer-30b $0.32 input · $1.26 output / 1M tokens Minimax m3 minimax/minimax-m3 $0.16 input · $0.63 output / 1M tokens Kimi k3 moonshotai/kimi-k3 $2.99 input · $14.96 output / 1M tokens Gpt Oss 20b openai/gpt-oss-20b $0.03 input · $0.14 output / 1M tokens Qwen3 Embedding 8B Qwen/Qwen3-Embedding-8B $0.01 input · output / 1M tokens Qwen3 Reranker 8B Qwen/Qwen3-Reranker-8B input · output / 1M tokens Inkling thinkingmachines/inkling $0.53 input · $2.13 output / 1M tokens