macrostack

AI API price index · open-model host · read 17 Nov 2026

Cloudflare Workers AI API pricing

47 prices as Cloudflare Workers AI publishes them — 40 text models, 7 embeddings — read on its own pricing page on 17 Nov 2026. Its current text models cost $0.112 to $6.667 per million output tokens.

Free tier Workers AI is included in both the Free and Paid Workers plans: "Our free allocation allows anyone to use a total of 10,000 Neurons per day at no charge." Above that, Workers Paid is charged "$0.011 / 1,000 Neurons"; on Workers Free, more than 10,000 Neurons a day needs an upgrade. Limits reset daily at 00:00 UTC.

What to know before you build on Cloudflare Workers AI

  • Everything is billed in Neurons at $0.011 per 1,000; the per-token prices are Cloudflare's own conversion of each model's Neuron rate.
  • 10,000 Neurons a day are free on both the Free and Paid Workers plans, reset at 00:00 UTC.
  • The cheapest text model is IBM's granite-4.0-h-micro, $0.017 in and $0.112 out per million tokens. gpt-oss-120b is $0.35 in and $0.75 out.
  • Some models, among them Kimi K2.6, GLM-5.3 and the DeepSeek V4 models, need a paid billing method.
  • Whisper speech recognition is $0.0005 per audio minute; it sits on the same page and is not ranked here.

Against the other 31

Across the index, 390 current text models run from $0.03 to $50.00 per million output tokens — 1667×. Cloudflare Workers AI’s cheapest, @cf/ibm-granite/granite-4.0-h-micro at $0.112, has 8 cheaper text models elsewhere; its dearest, @cf/meta/llama-2-7b-chat-fp16, costs $6.667. A cheaper model is a different model, not the same one for less — compare capability before price.

The same model elsewhere

Open-weight models Cloudflare Workers AI sells that other hosts price too — the only like-for-like comparison in the index.

  • gpt-oss-120b — Novita AI $0.05 / $0.25 · Google Vertex AI (Agent Platform) $0.09 / $0.36 · SambaNova Cloud $0.22 / $0.59 · Amazon Bedrock $0.15 / $0.60 · Fireworks AI $0.15 / $0.60 · Groq $0.15 / $0.60 · Nebius Token Factory $0.15 / $0.60 · Together AI $0.15 / $0.60 · Cerebras $0.35 / $0.75 · Cloudflare Workers AI $0.35 / $0.75 — 3.0× apart
  • Qwen3.8 27B — Cerebras $0.99 / $1.49 · DeepInfra $0.20 / $2.50 · Alibaba Cloud Model Studio (Qwen) $0.50 / $3.00 · Nebius Token Factory $0.45 / $3.00 · Novita AI $0.42 / $3.00 · Cloudflare Workers AI $0.45 / $3.20 · Groq $0.80 / $4.00 — 2.7× apart
  • GLM 5.3 — DeepInfra $0.90 / $4.00 · Cloudflare Workers AI $1.40 / $4.40 · Fireworks AI $1.40 / $4.40 · Mistral AI $1.40 / $4.40 · Nebius Token Factory $1.40 / $4.40 · Novita AI $1.40 / $4.40 · Perplexity $1.40 / $4.40 · Together AI $1.40 / $4.40 · Z.ai $1.40 / $4.40 · SambaNova Cloud $2.80 / $8.80 — 2.2× apart
  • DeepSeek V4.1 Flash 1017 — DeepInfra $0.20 / $0.60 · Cloudflare Workers AI $0.30 / $1.20 · DeepSeek $0.30 / $1.20 · Nebius Token Factory $0.30 / $1.20 · Novita AI $0.30 / $1.20 — 2.0× apart
  • GLM 5.2 — DeepInfra $0.75 / $2.40 · Cloudflare Workers AI $1.40 / $4.40 · Google Vertex AI (Agent Platform) $1.40 / $4.40 · Nebius Token Factory $1.40 / $4.40 · Novita AI $1.40 / $4.40 · Together AI $1.40 / $4.40 · Z.ai $1.40 / $4.40 — 1.8× apart
  • DeepSeek V4 Pro 0813 — DeepInfra $1.30 / $2.60 · Cloudflare Workers AI $1.32 / $3.96 · DeepSeek $1.32 / $3.96 · Nebius Token Factory $1.32 / $3.96 · Novita AI $1.32 / $3.96 · Perplexity $1.32 / $3.96 · Together AI $1.32 / $3.96 — 1.5× apart
  • gpt-oss-20b — Google Vertex AI (Agent Platform) $0.07 / $0.25 · Amazon Bedrock $0.07 / $0.30 · Cloudflare Workers AI $0.20 / $0.30 · Groq $0.075 / $0.30 — 1.2× apart
  • GLM 5.3 Flash — Cloudflare Workers AI $0.15 / $0.50 · DeepInfra $0.15 / $0.50 · Fireworks AI $0.15 / $0.50 · Nebius Token Factory $0.15 / $0.50 · Novita AI $0.15 / $0.50 · Perplexity $0.15 / $0.50 · Together AI $0.15 / $0.50 · Z.ai $0.15 / $0.50 — the same everywhere

Input / output per 1M tokens, standard tier.

Text models

ModelInputCached inputOutputNote
@cf/meta/llama-3.2-1b-instruct$0.027—$0.201—
@cf/meta/llama-3.2-3b-instruct$0.051—$0.335—
@cf/meta/llama-3.1-8b-instruct-fp8-fast$0.045—$0.384—
@cf/meta/llama-3.2-11b-vision-instruct$0.049—$0.676—
@cf/meta/llama-3.1-70b-instruct-fp8-fast$0.293—$2.253—
@cf/meta/llama-3.3-70b-instruct-fp8-fast$0.293—$2.253—
@cf/deepseek-ai/deepseek-r1-distill-qwen-32b$0.497—$4.881—
@cf/deepseek-ai/deepseek-v4-flash-0731$0.44$0.014$1.32—
@cf/deepseek-ai/deepseek-v4.1-flash-1017$0.30$0.006$1.20—
@cf/deepseek-ai/deepseek-v4-pro-0813$1.32$0.044$3.96—
@cf/mistral/mistral-7b-instruct-v0.1$0.11—$0.19—
@cf/mistralai/mistral-small-3.1-24b-instruct$0.351—$0.555—
@cf/meta/llama-3.1-8b-instruct$0.282—$0.827—
@cf/meta/llama-3.1-8b-instruct-fp8$0.152—$0.287—
@cf/meta/llama-3.1-8b-instruct-awq$0.123—$0.266—
@cf/meta/llama-3-8b-instruct$0.282—$0.827—
@cf/meta/llama-3-8b-instruct-awq$0.123—$0.266—
@cf/meta/llama-2-7b-chat-fp16$0.556—$6.667—
@cf/meta/llama-guard-3-8b$0.484—$0.03Not ranked: a safety classifier, not a general model.
@cf/meta/llama-4-scout-17b-16e-instruct$0.27—$0.85—
@cf/google/gemma-3-12b-it$0.345—$0.556—
@cf/qwen/qwq-32b$0.66—$1.00—
@cf/qwen/qwen2.5-coder-32b-instruct$0.66—$1.00—
@cf/qwen/qwen3-30b-a3b-fp8$0.051—$0.335—
@cf/qwen/qwen3.8-27b$0.45—$3.20—
@cf/openai/gpt-oss-120b$0.35—$0.75—
@cf/openai/gpt-oss-20b$0.20—$0.30—
@cf/aisingapore/gemma-sea-lion-v4-27b-it$0.351—$0.555—
@cf/ibm-granite/granite-4.0-h-micro$0.017—$0.112—
@cf/zai-org/glm-4.7-flash$0.06—$0.40—
@cf/zai-org/glm-5.2$1.40$0.26$4.40—
@cf/zai-org/glm-5.3$1.40$0.26$4.40—
@cf/zai-org/glm-5.3-flash$0.15$0.03$0.50—
@cf/nvidia/nemotron-3-120b-a12b$0.50—$1.50—
@cf/moonshotai/kimi-k2.5$0.60$0.10$3.00—
@cf/moonshotai/kimi-k2.6$0.95$0.16$4.00—
@cf/moonshotai/kimi-k2.7-code$0.95$0.19$4.00—
@cf/google/gemma-4-26b-a4b-it$0.10—$0.30—
@cf/daily-co/phonellm-mini-alpha-2$0.06—$0.24—
@cf/thinkingmachines/inkling-2-small$0.30$0.10$1.00—

Embeddings

ModelInputCached inputNote
@cf/baai/bge-small-en-v1.5$0.02——
@cf/baai/bge-base-en-v1.5$0.067——
@cf/baai/bge-large-en-v1.5$0.204——
@cf/baai/bge-m3$0.012——
@cf/pfnet/plamo-embedding-1b$0.019——
@cf/qwen/qwen3-embedding-0.6b$0.012——
@cf/google/embeddinggemma-2$0.0017——

46 of these 47 prices are ranked in the index; the rest say why in their note. Text and embeddings per 1M tokens unless the row says otherwise.

Discounts and surcharges, as published

  • Everything is billed in Neurons at $0.011 per 1,000; the per-token prices are Cloudflare's own conversion of each model's Neuron rate ("The Price in Tokens column is equivalent to the Price in Neurons column").
  • Prepaid AI Gateway credits can pay for Workers AI inference; "Requests to frontier models that use prepaid credits receive higher rate limits."
  • Some models require a paid billing method, among them the Kimi K2.6 and K2.7 Code, GLM-5.2, GLM-5.3, GLM-5.3-Flash and DeepSeek V4 models.

Where these numbers come from

Read on developers.cloudflare.com/workers-ai/platform/pricing on 17 Nov 2026, checked against 1 more of Cloudflare Workers AI’s own pages. Macrostack measures nothing and is paid by no provider on this index. Prices change; the date is the claim.

The other providers in the index

Questions

How much does the Cloudflare Workers AI API cost?
On 17 Nov 2026, Cloudflare Workers AI's current text models cost $0.112 to $6.667 per million output tokens at the standard tier (@cf/ibm-granite/granite-4.0-h-micro to @cf/meta/llama-2-7b-chat-fp16). The full list, with input and cached-input prices, is on this page.
Does Cloudflare Workers AI have a free tier?
What Cloudflare Workers AI's own pages say: Workers AI is included in both the Free and Paid Workers plans: "Our free allocation allows anyone to use a total of 10,000 Neurons per day at no charge." Above that, Workers Paid is charged "$0.011 / 1,000 Neurons"; on Workers Free, more than 10,000 Neurons a day needs an upgrade. Limits reset daily at 00:00 UTC.
Does Cloudflare Workers AI offer discounts?
Everything is billed in Neurons at $0.011 per 1,000; the per-token prices are Cloudflare's own conversion of each model's Neuron rate ("The Price in Tokens column is equivalent to the Price in Neurons column"). Prepaid AI Gateway credits can pay for Workers AI inference; "Requests to frontier models that use prepaid credits receive higher rate limits." Some models require a paid billing method, among them the Kimi K2.6 and K2.7 Code, GLM-5.2, GLM-5.3, GLM-5.3-Flash and DeepSeek V4 models.
The Macrostack brief

New swaps, worth your inbox.

A short, occasional email when we add a high-intent alternative or ship a new head-to-head. No spam, no selling your address — unsubscribe in one click.