Cohere
Enterprise AI with RAG and embeddings focus
RAGEmbeddingsEnterpriseMultilingualOpenAI-compatible
About Cohere
Cohere builds AI models optimized for enterprise use cases including retrieval-augmented generation (RAG), semantic search, and text classification. Their Command model family is designed for production deployments with strong multilingual support.
Key Models
Command ACommand R+Embed v3
Live Performance Data
Cohere Models on Antbase
| Model | Score | Requests | Success Rate | Avg Latency | p95 Latency | Capabilities |
|---|---|---|---|---|---|---|
| Rerank v3.5 | 80.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed Multilingual v3.0 | 75.0 | 0 | 0.0% | 400ms | 1000ms | |
| Rerank Multilingual v3.0 | 75.0 | 0 | 0.0% | 400ms | 1000ms | |
| Rerank English v3.0 | 75.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed English v3.0 | 75.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed Multilingual Light v3.0 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed English Light v3.0 | 60.0 | 0 | 0.0% | 400ms | 1000ms | |
| Tiny Aya Water | 50.0 | 0 | 0.0% | 400ms | 1000ms | |
| C4ai Aya Expanse 8b | 50.0 | 0 | 0.0% | 400ms | 1000ms | |
| Command A Vision 07 2025 | 50.0 | 0 | 0.0% | 400ms | 1000ms | 👁🔧 |
| C4ai Aya Vision 8b | 50.0 | 0 | 0.0% | 400ms | 1000ms | 👁 |
| C4ai Aya Expanse 32b | 50.0 | 0 | 0.0% | 400ms | 1000ms | |
| Command R7b Arabic 02 2025 | 50.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Command A Translate 08 2025 | 50.0 | 0 | 0.0% | 400ms | 1000ms | 🔧 |
| Tiny Aya Earth | 50.0 | 0 | 0.0% | 400ms | 1000ms | |
| Tiny Aya Fire | 50.0 | 0 | 0.0% | 400ms | 1000ms | |
| Command A Reasoning 08 2025 | 50.0 | 0 | 0.0% | 400ms | 1000ms | 🧠🔧 |
| Tiny Aya Global | 50.0 | 0 | 0.0% | 400ms | 1000ms | |
| C4ai Aya Vision 32b | 50.0 | 0 | 0.0% | 400ms | 1000ms | 👁 |
| Embed Multilingual V3.0 Image | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Rerank V4.0 Fast | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed Multilingual Light V3.0 Image | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed English V2.0 | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed Multilingual V2.0 | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed English Light V3.0 Image | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed English V3.0 Image | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Rerank V4.0 Pro | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed English Light V2.0 | 30.0 | 0 | 0.0% | 400ms | 1000ms | |
| Embed V4.0 | 30.0 | 0 | 0.0% | 400ms | 1000ms |
How to Use Cohere on Antbase
All Cohere models are available through Antbase's unified OpenAI-compatible API. You can either route directly to a specific Cohere model or let ANT's intelligent router automatically select the best model for your task.
python
from openai import OpenAI
client = OpenAI(
base_url="https://antbase.ai/v1",
api_key="your-ant-api-key",
)
# Let ANT auto-route (may use Cohere if optimal)
response = client.chat.completions.create(
model="ant:auto",
messages=[{"role": "user", "content": "Hello!"}],
)
# Or route directly to a Cohere model
response = client.chat.completions.create(
model="cohere/command-a",
messages=[{"role": "user", "content": "Hello!"}],
)Why Use Cohere Through Antbase?
- •No separate account needed — access Cohere models instantly through your Antbase API key.
- •Intelligent routing — ANT automatically selects the best Cohere model for each task, or falls over to another provider if Cohere is slow or down.
- •Same token prices — Antbase charges zero markup on Cohere token prices. You pay exactly what Cohere charges.
- •Automatic failover — if Cohere has an outage, your requests seamlessly route to an equivalent model on another provider.