LLM Reference
NVIDIA NIM

Models on NVIDIA NIM

139 models available · NVIDIA

ModelInput (per 1M)Output (per 1M)Context
ArcticFreeFree4K
Baichuan 2 13B ChatFreeFree
Bielik 11B v2.6 InstructFreeFree4K
Breeze 7BFreeFree
ChatGLM3 6BFreeFree8K
CodeGemma 1.1 7BFreeFree
CodeGemma 7B InstructFreeFree
CodeLlama 70BFreeFree16K
Codestral 22BFreeFree32K
Codestral Mamba 7BFreeFree256K
Colosseum 355B InstructFreeFree16K
DBRX InstructFreeFree32K
DeepSeek Coder 6.7BFreeFree
DeepSeek R1FreeFree128K
DeepSeek R1 Distill Llama 8BFreeFree128K
DeepSeek R1 Distill Qwen 14BFreeFree128K
DeepSeek R1 Distill Qwen 32BFreeFree128K
DeepSeek R1 Distill Qwen 7BFreeFree128K
DeepSeek V3FreeFree64k
DeepSeek V3.1FreeFree33K
DeepSeek V3.1 TerminusFreeFree164K
DeepSeek V3.2FreeFree160K
DePlotFreeFree
Dracarys Llama 3.1 70B InstructFreeFree8K
Falcon 3 7B InstructFreeFree128K
Fuyu-8BFreeFree
Gemma 2 27B InstructFreeFree8K
Gemma 2 2B InstructFreeFree
Gemma 2 9B InstructFreeFree8K
Gemma 2 9B SahabatAI InstructFreeFree8K
Gemma 2B InstructFreeFree2K
Gemma 3 1B InstructFreeFree32K
Gemma 3 27B (free)FreeFree131K
Gemma 3n 2B (free)FreeFree8K
Gemma 3n 4B (free)FreeFree8K
Gemma 7B InstructFreeFree8K
GLM-4.7FreeFree
GLM-5FreeFree
gpt-oss-120bFreeFree131072
gpt-oss-20bFreeFree131072
Granite 3.3 8B InstructFreeFree128K
Granite 34B CodeFreeFree8K
Granite 8B CodeFreeFree8K
Granite Guardian 3.0 8BFreeFree8K
Italia 10B InstructFreeFree16K
Jamba 1.5 MiniFreeFree256K
Kimi K2 InstructFreeFree
Kimi K2 Instruct 0905FreeFree256K
Kimi K2 ThinkingFreeFree256K
Kimi K2.5FreeFree256K
Kosmos 2FreeFree
Llama 2 70B ChatFreeFree4K
Llama 3 70B InstructFreeFree8K
Llama 3 8B InstructFreeFree8K
Llama 3 Swallow 70B InstructFreeFree4K
Llama 3 Taiwan 70B InstructFreeFree8K
Llama 3.1 405B InstructFreeFree128K
Llama 3.1 70B InstructFreeFree128K
Llama 3.1 8B InstructFreeFree128K
Llama 3.1 NemoGuard 8B Content SafetyFreeFree4K
Llama 3.1 NemoGuard 8B Topic ControlFreeFree4K
Llama 3.1 Nemotron 70B RewardFreeFree4K
Llama 3.1 Nemotron Nano 4B v1.1FreeFree4K
Llama 3.1 Nemotron Nano 8B v1FreeFree4K
Llama 3.1 Nemotron Nano VL 8B v1FreeFree4K
Llama 3.1 Swallow 70B InstructFreeFree4K
Llama 3.1 Swallow 8B InstructFreeFree4K
Llama 3.2 11B Vision InstructFreeFree128K
Llama 3.2 1B InstructFreeFree128K
Llama 3.2 3B InstructFreeFree128K
Llama 3.2 90B Vision InstructFreeFree128K
Llama 3.2 NV EmbedQA 1B v2FreeFree4K
Llama 3.2 NV RerankQA 1B v2FreeFree4K
Llama 3.3 70B Instruct (free)FreeFree66K
Llama 3.3 Nemotron Super 49B v1FreeFree128K
Llama 4 Maverick 17B Instruct FP8FreeFree1M
Llama 4 Scout 17B-16E InstructFreeFree328K
Llama Guard 4 12BFreeFree164K
LLaVA 1.6 Hermes Yi 34BFreeFree200K
LLaVA 1.6 Mistral 7BFreeFree32K
Magistral Small 2506FreeFree128K
Marin 7B InstructFreeFree8192
Marin 8B InstructFreeFree128K
MiniMax M2.5FreeFree197K
Mistral 7B Instruct v0.2FreeFree32K
Mistral 7B Instruct v0.3FreeFree32K
Mistral 7B v0.1FreeFree8K
Mistral LargeFreeFree32k
Mistral Large 3 675B InstructFreeFree128K
Mistral Medium 3 InstructFreeFree128K
Mistral NeMo Instruct (2407)FreeFree128K
Mistral NemotronFreeFree
Mistral Small 3.1 24B InstructFreeFree128K
Mistral Small 4FreeFree256k
Mixtral 8x22B v0.1FreeFree64K
Mixtral 8x7BFreeFree32K
Nemotron 3 NanoFreeFree256K
Nemotron 3 Super-120B-A12BFreeFree1M
Nemotron 4 340BFreeFree4K
Nemotron Mini 4B InstructFreeFree4K
Nemotron Mini Hindi 4B InstructFreeFree4K
Nemotron-Nano-12B-v2-VLFreeFree
Nemotron-Nano-9B-v2FreeFree
NeVA 22BFreeFree
NV-EmbedCode 7B v1FreeFree4K
NVIDIA Llama 3 ChatQA 70BFreeFree
NVIDIA Llama 3 ChatQA 8BFreeFree
PaliGemma 3B 896FreeFree512
Phi 3.5 Mini InstructFreeFree128K
Phi 4 Multimodal InstructFreeFree128K
Phi-3 Medium 128KFreeFree128K
Phi-3 Medium 4KFreeFree4K
Phi-3 Mini 128KFreeFree128K
Phi-3 Mini 4kFreeFree4K
Phi-3 Small 128KFreeFree128K
Phi-3 Small 8KFreeFree8K
Phi-3 VisionFreeFree128K
Phi-4 MiniFreeFree
Phi-4 Mini Flash ReasoningFreeFree128K
Qwen2 7BFreeFree128K
Qwen2 7B InstructFreeFree128K
Qwen2.5 7B InstructFreeFree128K
Qwen2.5 Coder 32B InstructFreeFree
Qwen2.5 Coder 7B InstructFreeFree
Qwen3 Coder 480B A35B InstructFreeFree256K
RakutenAI 7B ChatFreeFree4K
RakutenAI 7B InstructFreeFree4K
RecurrentGemma 2BFreeFree
Sarvam-M Multilingual HybridFreeFree128K
SEA-LION 7BFreeFree
SeaLLM 7B V2.5FreeFree
Seed-OSS 36B InstructFreeFree4K
ShieldGemma 9BFreeFree8K
SOLAR 10.7BFreeFree
StarCoder2 15BFreeFree8K
StarCoder2 7BFreeFree8K
Stockmark 2 100B InstructFreeFree128K
Teuken 7B InstructFreeFree4K
Yi LargeFreeFree32K

About NVIDIA NIM

NVIDIA's AI platform offers a comprehensive ecosystem for developing, deploying, and scaling AI applications across industries. At its core, the platform leverages GPU-accelerated computing to enhance deep learning and machine learning workloads. It includes NVIDIA AI Enterprise, providing cloud-native tools for data science and generative AI, and NVIDIA NIM (Inference Microservices) for rapid deployment of production AI models. The platform supports various AI frameworks and libraries, ensuring optimized performance for diverse AI workloads and reducing application launch times from weeks to minutes. The platform's infrastructure seamlessly integrates on-premises and cloud environments, featuring NVIDIA DGX and HGX platforms for high-performance computing and AI tasks. This robust infrastructure is designed with energy efficiency in mind, allowing for innovation without excessive operational costs. The AI Accelerated program showcases validated applications that utilize this powerful infrastructure, enabling organizations to fast-track their AI initiatives while maintaining security and scalability across operations. This comprehensive approach allows customers to efficiently manage complex AI workflows and accelerate their AI-driven projects.

Full provider profile →