LLM API cost calculator

Enter your monthly token volume and see what the same model costs at every provider that serves it, at their advertised list prices.

Llama 3.3 70B Instruct: 10M input + 2M output tokens per month

#	Provider	Type	Estimated / month
1	DeepInfra	Inference hosts	$1.64	✓
2	OpenRouter	LLM routers	$1.64	✓
3	Inference.net	Inference hosts	$2.10	✓
4	Requesty	LLM routers	$2.10	✓
5	Novita AI	Inference hosts	$2.15	✓
6	AkashML	Inference hosts	$3.04	✓
7	Parasail	Inference hosts	$3.20	✓
8	GMI Cloud Inference Engine	Inference hosts	$4	✓
9	Hyperbolic	Inference hosts	$4.80	◐
10	Cloudflare Workers AI	Inference hosts	$7.44	✓
11	Featherless	Inference hosts	$8	✓
12	SambaNova Cloud (SambaCloud)	Inference hosts	$8.40	✓
13	CoreWeave	Inference hosts	$8.52	✓
14	Microsoft Foundry (Azure AI Foundry / Azure OpenAI)	Hyperscaler AI platforms	$8.52	✓
15	Amazon Bedrock	Hyperscaler AI platforms	$8.64	✓
16	Google Vertex AI (Gemini Enterprise Agent Platform)	Hyperscaler AI platforms	$8.64	✓
17	IBM watsonx.ai	Hyperscaler AI platforms	$9.03	✓
18	OVHcloud AI Endpoints	Inference hosts	$9.23	✓
19	Scaleway	Inference hosts	$12.40	✓
20	Together AI	Inference hosts	$12.48	✓
21	Venice.ai API	Inference hosts	$12.60	✓
22	Groq	Inference hosts	—	◐

List prices only: no volume discounts, batch pricing, taxes or router platform fees. Cache hits use the provider's published cached-input price. Where none is published they are billed at the normal input price.