Use NVIDIA Models in Big-AGI.

DeepSeek flagship reasoning MoE (official 0813 release) with the full 1M context. Can be slow to cold-start on the free endpoint.

Fastest Nemotron MoE (30B, 3B active), 1M context, reasoning and tool use. Text only.

Meta open multimodal reasoning model (30B dense) distilled from Muse Spark, with tool calling. Always reasons.

Fast DeepSeek V4 MoE (284B, 13B active) with the full 1M context and reasoning.

Translation-specialized model (37 languages, few-shot prompting), 8K context.

NVIDIA quantum-calibration VLM, preview (domain-specific, not for general chat).

Thinking Machines reasoning model (preview). Can be unstable under load on the free endpoint. Retires on NVIDIA 2026-08-24.

Moonshot flagship MoE, natively multimodal (image inputs), with the full 1M context and reasoning. Currently degraded on NVIDIA.

Poolside compact coding-focused reasoning model. Can be slow to cold-start on the free endpoint.

Zhipu GLM 5.2 reasoning model. NVIDIA serves a reduced 202K context (native: 1M). Retires on NVIDIA 2026-08-24.

Experimental diffusion language model. CAUTION: prone to hanging under load.

NVIDIA flagship open hybrid Mamba-Transformer MoE (550B, 55B active), 1M context, reasoning and tool use.

NVIDIA content-safety classifier (not a chat model).

MiniMax M3 reasoning model with image inputs. NVIDIA serves a reduced 256K context (native: 1M). Retires on NVIDIA 2026-09-08.

StepFun fast multimodal reasoning model, 256K context.

Omni-modal Nemotron Nano (image, video and audio inputs), 1M context, reasoning.

Mistral frontier-class multimodal model with adjustable reasoning.

Fast DeepSeek V4 MoE (284B) with 1M context and reasoning.

DeepSeek flagship reasoning MoE. NVIDIA serves a reduced 256K context (native: 1M). Note: often slow or saturated on the free endpoint.

NVIDIA quantum-calibration VLM (domain-specific, not for general chat).

Google open multimodal model. CAUTION: NVIDIA serves 131K context and silently truncates longer inputs.

Open hybrid Mamba-Transformer MoE (120B, 12B active), 1M context, reasoning and tool use.

Qwen 3.5 flagship MoE (multimodal, reasoning). CAUTION: served only to some NVIDIA accounts - most keys get a 404 "Function not found for account" error.

Efficient open MoE (30B, 3B active) for high-volume tasks, 1M context, reasoning and tool use. Retires on NVIDIA 2026-08-31.

Translation-specialized model, 8K context.

NVIDIA content-safety classifier (not a chat model).

Vision-language Nemotron Nano for image and video understanding (verified image input). Retires on NVIDIA 2026-08-25.

Llama 3.3 70B distilled and post-trained by NVIDIA for reasoning and agentic tasks. Retires on NVIDIA 2026-08-25.

Small hybrid Mamba-Transformer for fast, cheap reasoning and tool use. Retires on NVIDIA 2026-08-25.

OpenAI open-weight MoE (117B, 5.1B active) with adjustable reasoning effort.

OpenAI open-weight MoE (20B, 3.6B active) with adjustable reasoning effort.

Mistral model post-trained by NVIDIA.

Small vision-language model, 16K context.

Meta content-safety classifier (not a chat model). NVIDIA serves a reduced 64K context.

Superseded by v1.5.

NVIDIA content-safety classifier (not a chat model).

NVIDIA topic-control guardrail (not a chat model).

Meta Llama 3.3 70B instruction-tuned, with tool calling. Retires on NVIDIA 2026-08-25 and is already unresponsive.

Llama large vision model. NVIDIA serves a reduced 32K context (native: 128K).

Llama vision model for image understanding.

Tiny Llama for edge-class tasks.

Small Llama for lightweight tasks.

Tiny legacy Nemotron, 4K context.

Meta Llama 3.1 70B instruction-tuned (superseded by Llama 3.3 70B).

Small fast Llama for utility tasks, with tool calling. Retires on NVIDIA 2026-08-25.