StatusOperational Live status
LLMAPI discountUp to −25%
Models on LLMAPI1
From$0.2 / 1M blended
RegionsUS, EU
Best uptime (24h)94.34%
Verdict
Groq designs custom LPU chips that deliver the lowest-latency open-model inference available, via a simple API. Through LLMAPI you reach 9 Groq 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
- Speech-to-Text
- Fastest Throughput
- Lowest Latency (TTFT)
- Best for High Volume Production
- Open-Weight Models
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 Groq, in one table.
Overview | |
| Provider | Groq |
| Website | groq.com |
| Best for | Ultra-low-latency inference for real-time apps |
| Models on LLMAPI | 1 |
| Popular models | GPT OSS 20B |
Access & routing | |
| API compatibility | OpenAI-compatible — /v1/chat/completions |
| Model ID prefix | groq/<model> |
| Regions | US, EU |
| 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 | $0.2 |
| Uptime (24h) | 94.34% (median across listed models) |
Modality | |
| Tags | |
Price & performance | |
| Tags | |
Deployment | |
| Tags | |
Trust & safety | |
| Tags | |
Privacy & data | |
| Inference location | US data centres |
| Data residency | Not documented by the provider |
| Default retention | Not retained after the request is served |
| Zero-data retention | Default Prompts are not stored by default |
| Training on API data | No |
| Human review | Not documented by the provider |
Compliance & lifecycle | |
| Certifications | SOC 2 · GDPR |
| 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 Groq models
A quick view of how Groq 's 9 models compare on intelligence, output speed and price — so you can pick the right one for your use case.
Most intelligent
#1Llama 4 Maverick40
#2Gemma 2 9B40
#3Llama 3.1 8B Instant39
#4Llama 3.3 70B38
Intelligence index · 9 models
Fastest
#1Gemma 2 9B940 t/s
#2Mixtral 8x7B655 t/s
#3Qwen3-32B546 t/s
#4Llama 4 Scout418 t/s
Output tokens / second · 9 models
Lowest price
#1Llama 3.3 70B$0.16
#2DeepSeek R1 Distill Llama 70B$1.58
#3Whisper Large v3$2.80
#4Llama 3.1 8B Instant$3.92
Blended price per 1M tokens · 9 models
Compared with other hosts
How Groq sits against the rest of the catalog, using the same measurements shown on every provider page.
- Price. Median blended price of $5.40 per 1M tokens — 12% below the $6.13 median across all 50 providers in the catalog.
- Throughput. Median output speed of 345 tokens/s, 1% slower than the 348 tok/s catalog median.
- Catalog depth. 9 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| Groq | $5.40 | |
| MiniMax | $5.41 | |
| Novita | $5.46 | |
| Fireworks | $5.50 | |
| Cohere | $5.50 | |
| Kimi (Moonshot AI) | $5.28 |
Median output speed
higher is better| Groq | 345 tok/s | |
| MiniMax | 273 tok/s | |
| Novita | 312 tok/s | |
| Fireworks | 480 tok/s | |
| Cohere | 310 tok/s | |
| Kimi (Moonshot AI) | 432 tok/s |
Figures are medians across the models listed on each provider page and are refreshed with the catalog.
Models & pricing
Every Groq model reachable with your LLMAPI key. Click a column heading to sort.
| Model | API model name | Uptime 24h | Price / 1M | Context |
|---|---|---|---|---|
| GPT OSS 20B | gpt-oss-20b | 94.34% | $ 0.2 | 131 k |
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.
Groq via LLMAPI
Same models, same features, one key. Prefix the model ID with groq/ 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="groq/llama-3-3-70b", 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 Groq’s own API. |
| Fallback | Configurable. On 429 or 5xx the router switches to any model or provider on your fallback list. |
| Regions | US, EU |
| 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 Groq models through LLMAPI?
Point any OpenAI-compatible client at https://api.llmapi.ai/v1 and use the model ID groq/llama-3-3-70b. No other change is needed.
Does Groq cost more through LLMAPI?
No. This provider is eligible for LLMAPI volume discounts, so heavy usage lands below list price.
Which regions are used?
US, EU
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 Groq models are available?
9 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.