LLM API Cost Calculator

Compare the monthly cost of 143+ LLM APIs from a single screen. Drag the sliders, see your exact bill.

Last updated 13 days ago

Pricing AI workloads is harder than it should be. Every provider quotes a different unit — per 1K tokens, per 1M tokens, per call, per second of audio — and most published rate cards leave out fixed costs, batching discounts, and image surcharges. This calculator normalises 409+ models from Railwail's catalogue so you can compare them apples-to-apples in seconds.

How the math works

Each model has a per-1K-token input price and a per-1K-token output price, stored in credits (1 credit = $0.01). Your monthly bill is simply input_tokens × input_price + output_tokens × output_price. We surface the per-1M numbers in the table because most providers quote that unit publicly, but the underlying math runs on per-1K. Fixed per-call costs (image generation, video, audio) don't apply to token-based models and are excluded from this view — see /pricing for per-call workloads.

Switch providers, keep the same code

All models in the table speak the same OpenAI-compatible API on Railwail. Once you've identified a cheaper option, swapping is one environment-variable change — no SDK rewrite, no retry logic. Compare any two models head-to-head on the /compare page, or browse the full catalogue at /models.

Switching from GPT-5.5 to OpenAI text-embedding-3-small would save you ~100% on a 1M-input + 1M-output workload — same monthly traffic, same prompts, just a different model behind the API.

Monthly input tokens100K

tokens / month

The prompt + context you send to the model each month.

Monthly output tokens100K

tokens / month

The total response volume the model generates.

Category

Token-based categories only — image, video, audio use per-call pricing.

Provider

Filter the table to one upstream provider.

Showing 143 models.Cheapest match: BLIP at $0.00030/month.

Sorted by monthly cost (asc)

					Action
BGE Large EN v1.5 huggingface	—	—	n/a	512	Try
BGE-M3 (Multilingual) huggingface	—	—	n/a	8K	Try
Bio_ClinicalBERT huggingface	—	—	n/a	—	Try
Biomedical NER (all entities) huggingface	—	—	n/a	—	Try
Claude 3.5 Sonnet (vision) Anthropic	—	—	n/a	200K	Try
Claude Opus 4 Anthropic New	—	—	n/a	200K	Try
Claude Sonnet 4 Anthropic	—	—	n/a	200K	Try
Codestral Mistral New	—	—	n/a	256K	Try
DeepSeek V3.1 DeepSeek	—	—	n/a	131K	Try
ESM-2 650M (Protein Embeddings) huggingface	—	—	n/a	1K	Try
Gemini 1.5 Pro (vision) Google	—	—	n/a	2.1M	Try
Gemini 2.0 Flash Google New	—	—	n/a	1.0M	Try
Grok 4.20 Reasoning Community	—	—	n/a	1.0M	Try
Grok 4.3 Community New	—	—	n/a	1.0M	Try
Kimi K2 (Moonshot) Community	—	—	n/a	131K	Try
Medical NER (DeBERTa) huggingface	—	—	n/a	—	Try
MiniMax-01 Community	—	—	n/a	4.1M	Try
Nomic Embed Text v1.5 huggingface	—	—	n/a	8K	Try
Perplexity Sonar Pro Community	—	—	n/a	200K	Try
PubMedBERT Embeddings (NeuML) huggingface	—	—	n/a	512	Try
SPECTER (Scientific Paper Embeddings) huggingface	—	—	n/a	512	Try
Voyage AI voyage-3 Community	—	—	n/a	32K	Try
AI21 Jamba 1.5 Large Community	—	—	n/a	256K	Try
AI21 Jamba 1.5 Mini Community	—	—	n/a	256K	Try
BLIP Image Captioning Large huggingface	—	—	n/a	—	Try
BioBERT Disease NER (NCBI) huggingface	—	—	n/a	—	Try
BioBERT v1.2 (Biomedical Embeddings) huggingface	—	—	n/a	512	Try
BiomedBERT (PubMedBERT abstract) huggingface	—	—	n/a	512	Try
Claude Haiku 3.5 Anthropic	—	—	n/a	200K	Try
Clinical Assertion and Negation BERT huggingface	—	—	n/a	—	Try
Clinical NER (problem, test, treatment) huggingface	—	—	n/a	—	Try
CodeGen 350M Mono huggingface	—	—	n/a	2K	Try
Cohere Aya 23 35B Community	—	—	n/a	8K	Try
Cohere Command Light (legacy) Community	—	—	n/a	4K	Try
Cohere Command R (08-2024) Community	—	—	n/a	128K	Try
Cohere Command R+ (08-2024) Community	—	—	n/a	128K	Try
Cohere embed-multilingual-v3 Community	—	—	n/a	512	Try
DeepSeek Coder 1.3B Instruct huggingface	—	—	n/a	16K	Try
DeepSeek Coder V2 DeepSeek	—	—	n/a	128K	Try
DeepSeek R1 DeepSeek New	—	—	n/a	64K	Try
DeepSeek V3 DeepSeek	—	—	n/a	64K	Try
GPT-6 Astra OpenAI New	—	—	n/a	1.1M	Try
GPT-6 Luna OpenAI New	—	—	n/a	1.1M	Try
GPT-6 Sol OpenAI New	—	—	n/a	1.1M	Try
GTE Large EN v1.5 huggingface	—	—	n/a	8K	Try
Gemini 1.5 Flash (vision) Google	—	—	n/a	1.0M	Try
Grok 4.20 (Non-Reasoning) Community	—	—	n/a	1.0M	Try
Grok 4.20 Multi-Agent Community	—	—	n/a	1.0M	Try
Grok Build 0.1 Community	—	—	n/a	256K	Try
Jina Embeddings v3 (Multilingual) Community	—	—	n/a	8K	Try
Llama 3.3 70B Together	—	—	n/a	131K	Try
Marker PDF Extract Replicate	—	—	n/a	—	Try
Mistral Large Mistral	—	—	n/a	128K	Try
Multilingual E5 Large huggingface	—	—	n/a	512	Try
Perplexity Sonar Community	—	—	n/a	127K	Try
Perplexity Sonar Reasoning Community	—	—	n/a	127K	Try
Qwen 2.5 72B Together	—	—	n/a	131K	Try
Qwen 2.5-Max Community	—	—	n/a	33K	Try
Qwen2.5-Coder 32B Instruct huggingface	—	—	n/a	131K	Try
Qwen2.5-Coder 7B Instruct huggingface	—	—	n/a	131K	Try
Qwen2.5-VL 7B Instruct (HF) huggingface	—	—	n/a	33K	Try
Reka Core Community	—	—	n/a	128K	Try
Reka Edge Community	—	—	n/a	16K	Try
Reka Flash Community	—	—	n/a	128K	Try
Replit Code v1.5 3B huggingface	—	—	n/a	4K	Try
SciBERT (scivocab uncased) huggingface	—	—	n/a	512	Try
Snowflake Arctic Instruct Community	—	—	n/a	4K	Try
Stable Code Instruct 3B huggingface	—	—	n/a	16K	Try
Voyage AI voyage-code-3 Community	—	—	n/a	32K	Try
Yi Large Community	—	—	n/a	33K	Try
mxbai-embed-large-v1 Community	—	—	n/a	512	Try
BLIP Replicate Cheapest	—	—	$0.0003	—	Try
MiDaS v3.1 Replicate Cheapest	—	—	$0.0003	—	Try
Segformer B5 Replicate Cheapest	—	—	$0.0003	—	Try
ZoeDepth Replicate	—	—	$0.0006	—	Try
Florence-2 Large Replicate	—	—	$0.0012	—	Try
Florence-2 Segmentation Replicate	—	—	$0.0012	—	Try
GOT-OCR 2.0 Replicate	—	—	$0.0012	—	Try
Idefics3 8B Replicate	—	—	$0.0012	8K	Try
OpenPose Replicate	—	—	$0.0012	—	Try
SeamlessM4T v2 Large (Text) Replicate	—	—	$0.0012	4K	Try
EasyOCR Replicate	—	—	$0.0014	—	Try
Grounded-SAM Replicate	—	—	$0.0015	—	Try
MiniCPM-V 2.6 Replicate	—	—	$0.0015	33K	Try
Moondream2 Replicate	—	—	$0.0020	—	Try
OpenAI text-embedding-3-small OpenAI	$0.024	—	$0.0024	8K	Try
Qwen2-VL 7B Instruct Replicate	—	—	$0.0024	—	Try
DeepSeek Coder 33B Instruct (GGUF) Replicate	—	—	$0.0030	16K	Try
Llama 3.2 Vision 11B (Ollama) Replicate	—	—	$0.0039	—	Try
Depth Anything v2 Replicate	—	—	$0.0050	—	Try
Code Llama 13B Instruct Replicate	—	—	$0.0065	16K	Try
Llama 3.2 Vision 90B Replicate	—	—	$0.0070	131K	Try
Code Llama 7B Instruct Replicate	—	—	$0.0074	16K	Try
DeepSeek-VL 7B Replicate	—	—	$0.0086	4K	Try
GLPN Depth Replicate	—	—	$0.0094	—	Try
CogVLM2 19B Replicate	—	—	$0.011	8K	Try
Dots OCR Replicate	—	—	$0.013	—	Try
OpenAI text-embedding-3-large OpenAI	$0.156	—	$0.016	8K	Try
SAM 2 (Segment Anything 2) Replicate	—	—	$0.018	—	Try
olmOCR Replicate	—	—	$0.020	—	Try
PaddleOCR v3 Replicate	—	—	$0.022	—	Try
WizardCoder 33B Replicate	—	—	$0.025	16K	Try
Mask2Former Replicate	—	—	$0.028	—	Try
Granite Code 8B Replicate	$0.060	$0.300	$0.036	128K	Try
Code Llama 34B Instruct Replicate	—	—	$0.041	16K	Try
Lotus-G Replicate	—	—	$0.041	—	Try
CLIP Interrogator Replicate	—	—	$0.046	—	Try
Code Llama 70B Instruct Replicate	—	—	$0.050	16K	Try
Molmo 7B Replicate	—	—	$0.050	—	Try
Donut Document Replicate	—	—	$0.065	—	Try
Granite Code 20B Replicate	$0.120	$0.600	$0.072	8K	Try
GPT-4o Mini OpenAI	$0.180	$0.720	$0.090	128K	Try
GPT-4o mini (vision) OpenAI	$0.180	$0.720	$0.090	128K	Try
Magicoder S CL 7B Replicate	—	—	$0.092	16K	Try
Marigold Replicate	—	—	$0.094	—	Try
LLaVA 1.6 Vicuna 13B Replicate	—	—	$0.102	4K	Try
GPT-5.4 Nano OpenAI New	$0.240	$1.50	$0.174	400K	Try
DeepSeek V4 Flash DeepSeek New	$0.360	$1.44	$0.180	1.0M	Try
DeepSeek V4.1 Flash DeepSeek New	$0.360	$1.44	$0.180	1.0M	Try
Phind CodeLlama 34B v2 Replicate	—	—	$0.216	16K	Try
GPT-5 Mini OpenAI	$0.300	$2.40	$0.270	400K	Try
Gemini 3 Flash Google New	$0.600	$3.60	$0.420	1.0M	Try
GPT-5.4 Mini OpenAI New	$0.900	$5.40	$0.630	400K	Try
DeepSeek V4 Pro DeepSeek New	$1.58	$4.75	$0.634	1.0M	Try
o3-mini OpenAI New	$1.32	$5.28	$0.660	200K	Try
OpenAI o4-mini OpenAI	$1.32	$5.28	$0.660	200K	Try
Claude Haiku 4.5 Anthropic New	$1.20	$6.00	$0.720	200K	Try
StarCoder2 15B Replicate	—	—	$1.19	16K	Try
GPT-4.1 OpenAI New	$2.40	$9.60	$1.20	1.0M	Try
OpenAI o3 OpenAI	$2.40	$9.60	$1.20	200K	Try
Gemini 2.5 Pro Google New	$1.50	$12.00	$1.35	1.0M	Try
GPT-5.1 OpenAI	$1.50	$12.00	$1.35	400K	Try
Claude Sonnet 5 Anthropic New	$2.40	$12.00	$1.44	1.0M	Try
GPT-4o OpenAI	$3.00	$12.00	$1.50	128K	Try
GPT-4o (vision) OpenAI	$3.00	$12.00	$1.50	128K	Try
Gemini 3.1 Pro Google New	$2.40	$14.40	$1.68	1.0M	Try
GPT-5.4 OpenAI New	$3.00	$18.00	$2.10	1.1M	Try
Claude Sonnet 4.6 Anthropic New	$3.60	$18.00	$2.16	1.0M	Try
Claude Opus 5.5 Anthropic New	$4.80	$24.00	$2.88	1.0M	Try
Claude Opus 4.7 Anthropic New	$6.00	$30.00	$3.60	1.0M	Try
Claude Opus 4.8 Anthropic New	$6.00	$30.00	$3.60	1.0M	Try
GPT-5.5 OpenAI New	$6.00	$36.00	$4.20	400K	Try
Claude Fable 5.1 Anthropic New	$12.00	$60.00	$7.20	1.0M	Try

Cheapest 5 models today

Ranked by combined input + output cost for a 1M + 1M token monthly workload.

OpenAI text-embedding-3-small

openai

$0.024 / mo

OpenAI text-embedding-3-large

Frequently asked questions

How is the monthly cost calculated?

For each model we apply (input_tokens × input_price + output_tokens × output_price) / 1000. Prices are stored in credits per 1K tokens (1 credit = $0.01); the table shows the USD per-1M equivalent because providers quote that unit publicly.

Why is OpenAI more expensive than open-weights alternatives?

Closed-frontier models bake R&D, alignment work, and hosted inference into a single price. Open-weights models (Llama, Mistral, DeepSeek) push the inference cost to hosting providers competing on margin, which is why per-token prices can be 5-50× lower for comparable quality.

What's the difference between input and output cost?

Input tokens are the prompt you send; output tokens are the model's response. Output is typically 2-5× more expensive because generation is sequential and dominates GPU time. Reasoning models (o1, DeepSeek R1) charge for hidden "thinking" tokens as output, which is the main cost driver.

Does this include batch discounts or volume pricing?

No — list price only. Most providers offer 25-50% discounts for batch endpoints and enterprise contracts; use the table as a ceiling, not a floor.

Sign up to start using these models

One OpenAI-compatible endpoint, every model in the table above. Free credits to start, transparent per-token pricing thereafter.

Start Building Browse All Models