Fireworks AI models and pricing

We track 12 models on Fireworks AI (27 listings) from 7 labs. The cheapest paid model is GLM-5.3-Flash, at $0.15 per million input tokens and $0.50 per million output tokens. Input prices range from $0.15 to $4.50 and output prices from $0.50 to $22.50 per million tokens. The largest context window is 1M tokens (Kimi K3.1 Pro Highspeed, Muse Spark Open and 2 other models). Prices last verified 18 Nov 2026.

Prices in USD per million tokens. Official site

Cheapest models on Fireworks AI

Input / output per million tokens, paid models, by combined price.

Newest models on Fireworks AI

By the lab's release date.

All Fireworks AI models and prices

ModelLabInputOutputContextSource
Kimi K3.1 Pro Highspeed
accounts/fireworks/models/kimi-k3p1-pro
Moonshot AI$2$101Mmodels.dev
MiniMax-M3.1-Flash
accounts/fireworks/models/minimax-m3p1-flash
MiniMax$0.15$0.601Mmodels.dev
DeepSeek V4.1 Flash 1017
accounts/fireworks/routers/deepseek-flash-latest
DeepSeek$0.30$1.201Mmodels.dev
DeepSeek V4.1 Flash 1017
accounts/fireworks/routers/deepseek-v4p1-flash-1017-ultrafast
DeepSeek$2.40$9.601Mmodels.dev
DeepSeek V4.1 Flash 1017
accounts/fireworks/models/deepseek-v4p1-flash-1017
DeepSeek$0.30$1.201Mmodels.dev
DeepSeek V4.1 Flash 1017
accounts/fireworks/routers/deepseek-v4p1-flash-1017-spot
DeepSeek$0.21$0.841Mmodels.dev
Muse Spark Open
accounts/fireworks/models/muse-spark-open
Meta$1.25$4.251Mmodels.dev
Ember-1
accounts/fireworks/models/ember-1
Fireworks$3$151Mmodels.dev
DeepSeek V4.1 Flash
accounts/fireworks/routers/deepseek-v4p1-flash-spot
DeepSeek$0.21$0.841Mmodels.dev
DeepSeek V4.1 Flash
accounts/fireworks/models/deepseek-v4p1-flash
DeepSeek$0.30$1.201Mmodels.dev
GLM-5.3-Flash
accounts/fireworks/models/glm-5p3-flash
Zhipu AI (Z.ai)$0.15$0.501Mmodels.dev
GLM-5.3-Flash
accounts/fireworks/routers/glm-flash-latest
Zhipu AI (Z.ai)$0.15$0.501Mmodels.dev
GLM-5.3
accounts/fireworks/routers/glm-fast-latest
Zhipu AI (Z.ai)$2.10$6.601Mmodels.dev
GLM-5.3
accounts/fireworks/models/glm-5p3
Zhipu AI (Z.ai)$1.40$4.401Mmodels.dev
GLM-5.3
accounts/fireworks/routers/glm-5p3-fast
Zhipu AI (Z.ai)$2.10$6.601Mmodels.dev
GLM-5.3
accounts/fireworks/routers/glm-latest
Zhipu AI (Z.ai)$1.40$4.401Mmodels.dev
GLM-5.3
accounts/fireworks/routers/glm-5p3-spot
Zhipu AI (Z.ai)$0.98$3.081Mmodels.dev
Qwen3.8 2.4T A95B
accounts/fireworks/models/qwen3p8-2p4t-a95b
Alibaba (Qwen)$2$6262Kmodels.dev
Qwen3.8 Max
accounts/fireworks/models/qwen3p8-max
Alibaba (Qwen)$2$6262Kmodels.dev
Qwen3.8 Max
accounts/fireworks/routers/qwen-max-latest
Alibaba (Qwen)$2$6262Kmodels.dev
Kimi K3
accounts/fireworks/routers/kimi-fast-latest
Moonshot AI$4.50$22.501Mmodels.dev
Kimi K3
accounts/fireworks/routers/kimi-k3-fast
Moonshot AI$4.50$22.501Mmodels.dev
Kimi K3
accounts/fireworks/routers/kimi-latest
Moonshot AI$2$101Mmodels.dev
Kimi K3
accounts/fireworks/models/kimi-k3
Moonshot AI$3$151Mmodels.dev
Kimi K3
accounts/fireworks/routers/kimi-k3-spot
Moonshot AI$2.10$10.501Mmodels.dev
MiniMax-M3
accounts/fireworks/routers/minimax-latest
MiniMax$0.30$1.20512Kmodels.dev
MiniMax-M3
accounts/fireworks/models/minimax-m3
MiniMax$0.30$1.20512Kmodels.dev

Frequently asked questions

How much does Fireworks AI cost per million tokens?

Prices on Fireworks AI depend on the model. Across the 12 models we track, input costs $0.15 to $4.50 per million tokens and output costs $0.50 to $22.50 per million tokens. Prices last verified 18 Nov 2026.

What is the cheapest model on Fireworks AI?

GLM-5.3-Flash from Zhipu AI (Z.ai) is the cheapest paid model we track on Fireworks AI, at $0.15 per million input tokens and $0.50 per million output tokens. Next are MiniMax-M3.1-Flash ($0.15 / $0.60) and DeepSeek V4.1 Flash ($0.21 / $0.84). Prices last verified 18 Nov 2026.

How many models does Fireworks AI offer?

We currently track 12 active models on Fireworks AI from 7 labs, led by Alibaba (Qwen) (2), DeepSeek (2), MiniMax (2) and Moonshot AI (2).

Which model on Fireworks AI has the largest context window?

Kimi K3.1 Pro Highspeed, Muse Spark Open and 2 other models share the largest context window we track on Fireworks AI: 1,048,576 tokens (1M).