StatusOperational Live status
LLMAPI discountUp to −25%
Models on LLMAPI3
From$1.01 / 1M blended
RegionsGlobal
Fastest model891 tok/s
Verdict
Wafer aggregates distributed GPU capacity into an affordable inference network for open models. Through LLMAPI you reach 3 Wafer models on the same OpenAI-compatible endpoint as every other provider, so switching between them is a one-string change.
Good fit for
- Text Generation
- Cheapest per Token
- Best Price-to-Performance
- Best for High Volume Production
- Open-Weight Models
- Best for Local Deployment
Consider another provider for
- No code generation models in this catalog
- No image generation models in this catalog
- No vision models in this catalog
- No video generation models in this catalog
Key facts
Where a provider does not document something we say so rather than guess. Everything LLMAPI records about Wafer, in one table.
Overview | |
| Provider | Wafer |
| Website | wafer.systems |
| Best for | Affordable open-model inference on distributed GPUs |
| Models on LLMAPI | 3 |
| Popular models | Llama 3.3 series, Mistral series, Qwen series |
Access & routing | |
| API compatibility | OpenAI-compatible — /v1/chat/completions |
| Model ID prefix | wafer/<model> |
| Regions | Global |
| Fallback | Configurable — on 429 or 5xx the router switches to any other model or provider on your list |
| Billing | One invoice across every provider, with per-key spend limits |
| Volume discount | Eligible |
Performance (median across listed models) | |
| Blended price / 1M | $6.22 |
Modality | |
| Tags | |
Price & performance | |
| Tags | |
Deployment | |
| Tags | |
Trust & safety | |
| Tags | |
Audience | |
| Tags | |
Privacy & data | |
| Inference location | Global |
| Data residency | Not documented by the provider |
| Default retention | Not documented by the provider |
| Zero-data retention | Not documented by the provider |
| Training on API data | Not documented by the provider |
| Human review | Not documented by the provider |
Compliance & lifecycle | |
| Certifications | SOC 2 · HIPAA |
| Deprecation notice | Not documented by the provider |
| Operational status | Operational Live status |
Prices and context from the OpenRouter public catalogue; uptime and output speed from OpenRouter provider endpoints; quality scores from LiveBench. Refreshed nightly — last updated 2026-09-29. — means no figure is published for that model.
Best Wafer models
A quick view of how Wafer 's 3 models compare on intelligence, output speed and price — so you can pick the right one for your use case.
Most intelligent
#1Qwen series37
#2Llama 3.3 series34
#3Mistral series21
Intelligence index · 3 models
Fastest
#1Llama 3.3 series891 t/s
#2Qwen series303 t/s
#3Mistral series281 t/s
Output tokens / second · 3 models
Lowest price
#1Llama 3.3 series$1.01
#2Qwen series$6.22
#3Mistral series$6.66
Blended price per 1M tokens · 3 models
Compared with other hosts
How Wafer sits against the rest of the catalog, using the same measurements shown on every provider page.
- Price. Median blended price of $6.22 per 1M tokens — 1% above the $6.13 median across all 50 providers in the catalog.
- Throughput. Median output speed of 303 tokens/s, 13% slower than the 348 tok/s catalog median.
- Catalog depth. 3 models listed, against a catalog average of 5 per provider.
- Integration. Identical to every other provider here — same endpoint, same SDK, same keys, so a switch costs one string change.
Median blended price per 1M tokens
lower is better| Wafer | $6.22 | |
| Alibaba Cloud | $6.23 | |
| SambaNova | $6.20 | |
| $6.31 | ||
| Arcee AI | $6.07 | |
| Anthropic | $6.63 |
Median output speed
higher is better| Wafer | 303 tok/s | |
| Alibaba Cloud | 150 tok/s | |
| SambaNova | 954 tok/s | |
| 239 tok/s | ||
| Arcee AI | 369 tok/s | |
| Anthropic | 365 tok/s |
Figures are medians across the models listed on each provider page and are refreshed with the catalog.
Models & pricing
Every Wafer model reachable with your LLMAPI key. Click a column heading to sort.
| Model | API model name |
|---|---|
| Llama 3.3 series | wafer/llama-3-3-series |
| Mistral series | wafer/mistral-series |
| Qwen series | wafer/qwen-series |
Price is a blended per-1M-token figure; latency is time to first token and throughput is output tokens per second, measured on the LLMAPI edge.
Cost calculator
Pick a model, enter your traffic, and see the monthly bill at LLMAPI rates.
Estimates use the blended per-1M-token price shown in the table above. Word counts differ by language — Cyrillic and CJK text uses 2–3× more tokens per word.
Wafer via LLMAPI
Same models, same features, one key. Prefix the model ID with wafer/ and point your OpenAI client at our base URL.
# Python · OpenAI SDK from openai import OpenAI client = OpenAI( base_url="https://api.llmapi.ai/v1", api_key="LLMAPI_KEY", ) r = client.chat.completions.create( model="wafer/llama-3-3-series", messages=[{"role": "user", "content": "Hello"}], )
| Pricing | Eligible for LLMAPI volume discounts; billed on one invoice with every other provider. |
| Features | Full pass-through — tool calling, JSON schema, streaming, vision and reasoning behave exactly as on Wafer’s own API. |
| Fallback | Configurable. On 429 or 5xx the router switches to any model or provider on your fallback list. |
| Regions | Global |
| Your data | LLMAPI stores request metadata only — model, token counts, latency and status. No prompt or completion content. |
Capabilities & use cases
Each tag below is its own catalog page listing every provider that shares it.
Developer resources
Where to go next.
Join thousands of developers building on one API key
Every provider in the catalog, one key, one invoice.
Questions
How do I call Wafer models through LLMAPI?
Point any OpenAI-compatible client at https://api.llmapi.ai/v1 and use the model ID wafer/llama-3-3-series. No other change is needed.
Does Wafer cost more through LLMAPI?
No. This provider is eligible for LLMAPI volume discounts, so heavy usage lands below list price.
Which regions are used?
Global
What happens if the provider returns an error?
On 429 or 5xx responses the router falls back to any other model or provider on your configured list, so requests keep succeeding.
How many Wafer models are available?
3 at the moment, all listed in the models table above and updated as the provider ships new ones.
Related providers
Providers covering similar use cases.