Skip to content

Models on kluster.ai＃

kluster.ai offers a wide variety of open-source models for both real-time and batch inferences, with more being constantly added.

This page covers all the models the API supports, with the API request limits for each.

Model names＃

Each model supported by kluster.ai has a unique name that must be used when defining the model in the request.

Model Model API name
DeepSeek R1 deepseek-ai/DeepSeek-R1
DeepSeek V3 deepseek-ai/DeepSeek-V3
DeepSeek V3 0324 deepseek-ai/DeepSeek-V3-0324
Gemma 3 27B google/gemma-3-27b-it
Llama 3.1 8B klusterai/Meta-Llama-3.1-8B-Instruct-Turbo
Llama 3.1 405B klusterai/Meta-Llama-3.1-405B-Instruct-Turbo
Llama 3.3 70B klusterai/Meta-Llama-3.3-70B-Instruct-Turbo
Qwen 2.5 7B Qwen/Qwen2.5-VL-7B-Instruct

Model comparison table＃

Model Main
use case		 Real-time
inference support		 Batch
inference support		 Fine-tuning
support		 Image
analysis		 Function
calling
DeepSeek R1 Code generation
Complex data analysis		 ✅ ✅ ❌ ❌ ❌
DeepSeek V3 Natural language generation
Contextually rich writing		 ✅ ✅ ❌ ❌ ❌
DeepSeek V3 0324 Natural language generation
Contextually rich writing		 ✅ ✅ ❌ ❌ ❌
Gemma 3 27B Multilingual applications
Image analysis
Complex reasoning		 ✅ ✅ ❌ ✅ ❌
Llama 3.1 8B Low-latency or simple tasks
Cost-efficient inference		 ✅ ✅ ✅ ❌ ✅
Llama 3.1 405B Detailed analysis
Maximum accuracy		 ✅ ✅ ❌ ❌ ✅
Llama 3.3 70B General-purpose AI
Balanced cost-performance		 ✅ ✅ ✅ ❌ ✅
Qwen 2.5 7B Document analysis
Image-based reasoning
Multimodal chat		 ✅ ✅ ❌ ✅ ❌

API request limits＃

The following limits apply to API requests based on your plan tier (notation is free tier | standard tier):

Model Context
size		 Max
output		 Max batch
requests		 Concurrent
requests		 Requests
per minute
DeepSeek R1 32k | 162k 4k | 162k <1000 | No limit 2 | 10 1 | 60
DeepSeek V3 32k | 131k 4k | 131k <1000 | No limit 2 | 10 1 | 60
DeepSeek V3 0324 32k | 131k 4k | 131k <1000 | No limit 2 | 10 1 | 60
Gemma 3 27B 32k | 32k 4k | 8k <1000 | No limit 2 | 10 1 | 60
Llama 3.1 8B 32k | 131k 4k | 131k <1000 | No limit 2 | 10 1 | 60
Llama 3.1 405B 32k | 131k 4k | 131k <1000 | No limit 2 | 10 1 | 60
Llama 3.3 70B 32k | 131k 4k | 131k <1000 | No limit 2 | 10 1 | 60
Qwen 2.5 7B 32k | 32k 4k | 8k <1000 | No limit 2 | 10 1 | 60