Big-AGI Logo
BIG-AGI

BEAM

Models

Features

Rankings

Docs

New

Pro

LAUNCH APP

NVIDIA NIM models in Big-AGI

Nemotron, GPT-OSS, DeepSeek, GLM - every model here costs $0 on one free nvapi- key. The catch is in the terms, not the invoice.

Nemotron 3.5 Super VL 120BStep 5 PreviewDeepSeek V4.1 FlashGLM 5.3DeepSeek V4 Pro 0813
NVIDIA setup guideNVIDIA models

Frontier open models, for $0.

Every model below runs at no cost on NVIDIA's hosted catalog: Nemotron 3, GPT-OSS, DeepSeek V4, GLM, MiniMax, Mistral, Gemma, and Llama. One free nvapi- key, no card. The throttle counts requests, not tokens, so long prompts and long answers go through where token-capped free tiers stop.

Two limits to know: expect roughly 40 requests per minute (NVIDIA publishes no number, and it moves with load), and NVIDIA's trial terms allow it to use your prompts to improve its models and exclude production use. Keep sensitive and production work on a paid provider.

All supported NVIDIA models

ModelContextInputOutputReleased
Nemotron 3.5 Super VL 120BNEWNemotron 3.5 Super early access: multimodal MoE (120B, 12B active) with image inputs, 1M context, reasoning and tool use.1MFreeFreeVisionReasoningTools / functionsOct 2026Step 5 PreviewNEWStepFun multimodal MoE preview (~600B, 27B active) with image inputs, tool calling and the full 1M context. Always reasons.1MFreeFreeVisionReasoningTools / functionsOct 2026DeepSeek V4.1 FlashHOTDeepSeek V4.1 multimodal MoE (552B, 8B active) with image inputs, tool calling, the full 1M context and reasoning.1MFreeFreeVisionReasoningTools / functionsSep 2026GLM 5.3 FlashZ.ai multimodal Flash on a new 320B MoE base (18B active), with image inputs, tool calling and the full 1M context. Always reasons.1MFreeFreeVisionReasoningTools / functionsAug 2026GLM 5.3deprecatedZ.ai text flagship (753B MoE) for coding and agentic work, with tool calling and the full 1M context. Always reasons. Retires on NVIDIA 2026-10-27.1MFreeFreeReasoningTools / functionsAug 2026DeepSeek V4 Pro 0813deprecatedDeepSeek flagship reasoning MoE (official 0813 release) with the full 1M context. Can be slow to cold-start on the free endpoint.1MFreeFreeReasoningTools / functionsAug 2026Nemotron 3.5 Lightning 30BFastest Nemotron MoE (30B, 3B active), 1M context, reasoning and tool use. Text only.1MFreeFreeReasoningTools / functionsAug 2026Muse Glimmer 30BMeta open multimodal reasoning model (30B dense) distilled from Muse Spark, with tool calling. Always reasons.131KFreeFreeVisionReasoningTools / functionsAug 2026DeepSeek V4 Flash 0731deprecatedFast DeepSeek V4 MoE (284B, 13B active) with the full 1M context and reasoning.1MFreeFreeReasoningTools / functionsJul 2026Riva Translate 4B v2Translation-specialized model (37 languages, few-shot prompting), 8K context.8KFreeFreeJul 2026Ising Calibration 1.5 31BNVIDIA quantum-calibration VLM, preview (domain-specific, not for general chat).131KFreeFreeVisionJul 2026InklingdeprecatedThinking Machines reasoning model (preview). Can be unstable under load on the free endpoint. Retires on NVIDIA 2026-08-24.131KFreeFreeVisionReasoningJul 2026Kimi K3deprecatedMoonshot flagship MoE, natively multimodal (image inputs), with the full 1M context and reasoning.1MFreeFreeVisionReasoningTools / functionsJul 2026Laguna XS 2.1Poolside compact coding-focused reasoning model. Can be slow to cold-start on the free endpoint.262KFreeFreeReasoningTools / functionsJul 2026GLM 5.2deprecatedZhipu GLM 5.2 reasoning model. NVIDIA serves a reduced 202K context (native: 1M). Retires on NVIDIA 2026-08-24.203KFreeFreeReasoningTools / functionsJun 2026DiffusionGemma 26BExperimental diffusion language model. CAUTION: prone to hanging under load.250KFreeFreeVisionReasoningTools / functionsJun 2026Nemotron 3 Ultra 550BNVIDIA flagship open hybrid Mamba-Transformer MoE (550B, 55B active), 1M context, reasoning and tool use.1MFreeFreeReasoningTools / functionsJun 2026Nemotron 3.5 Content SafetyNVIDIA content-safety classifier (not a chat model).131KFreeFreeJun 2026MiniMax M3deprecatedMiniMax M3 reasoning model with image inputs. NVIDIA serves a reduced 256K context (native: 1M). Retires on NVIDIA 2026-09-08.262KFreeFreeVisionReasoningTools / functionsMay 2026Step 3.7 FlashdeprecatedStepFun fast multimodal reasoning model, 256K context.262KFreeFreeVisionReasoningTools / functionsMay 2026Nemotron 3 Nano Omni 30BOmni-modal Nemotron Nano (image, video and audio inputs), 1M context, reasoning.1MFreeFreeVisionReasoningTools / functionsApr 2026Mistral Medium 3.5deprecatedMistral frontier-class multimodal model with adjustable reasoning.262KFreeFreeVisionReasoningTools / functionsApr 2026Ising Calibration 1 35BdeprecatedNVIDIA quantum-calibration VLM (domain-specific, not for general chat).262KFreeFreeVisionReasoningApr 2026Gemma 4 31BHOTGoogle open multimodal model. CAUTION: NVIDIA serves 131K context and silently truncates longer inputs.131KFreeFreeVisionReasoningTools / functionsApr 2026Nemotron 3 Super 120BOpen hybrid Mamba-Transformer MoE (120B, 12B active), 1M context, reasoning and tool use.1MFreeFreeReasoningTools / functionsMar 2026Qwen 3.5 397BdeprecatedQwen 3.5 flagship MoE (multimodal, reasoning). CAUTION: served only to some NVIDIA accounts - most keys get a 404 "Function not found for account" error.262KFreeFreeVisionReasoningTools / functionsFeb 2026Nemotron 3 Nano 30BdeprecatedEfficient open MoE (30B, 3B active) for high-volume tasks, 1M context, reasoning and tool use. Retires on NVIDIA 2026-08-31.1MFreeFreeReasoningTools / functionsDec 2025Riva Translate 4B v1.1deprecatedTranslation-specialized model, 8K context.8KFreeFreeDec 2025Nemotron Safety Guard 8B v3NVIDIA content-safety classifier (not a chat model).131KFreeFreeOct 2025Nemotron Nano 12B v2 VLdeprecatedVision-language Nemotron Nano for image and video understanding (verified image input). Retires on NVIDIA 2026-08-25.131KFreeFreeVisionReasoningTools / functionsOct 2025Llama 3.3 Nemotron Super 49B v1.5deprecatedLlama 3.3 70B distilled and post-trained by NVIDIA for reasoning and agentic tasks. Retires on NVIDIA 2026-08-25.131KFreeFreeReasoningTools / functionsOct 2025Nemotron Nano 9B v2deprecatedSmall hybrid Mamba-Transformer for fast, cheap reasoning and tool use. Retires on NVIDIA 2026-08-25.128KFreeFreeReasoningTools / functionsAug 2025GPT-OSS 20BOpenAI open-weight MoE (20B, 3.6B active) with adjustable reasoning effort.131KFreeFreeReasoningTools / functionsAug 2025GPT-OSS 120BdeprecatedOpenAI open-weight MoE (117B, 5.1B active) with adjustable reasoning effort.131KFreeFreeReasoningTools / functionsAug 2025Mistral NemotrondeprecatedMistral model post-trained by NVIDIA.262KFreeFreeReasoningTools / functionsJun 2025Nemotron Nano VL 8BdeprecatedSmall vision-language model, 16K context.16KFreeFreeVisionMay 2025Llama Guard 4 12BMeta content-safety classifier (not a chat model). NVIDIA serves a reduced 64K context.66KFreeFreeVisionApr 2025Llama 3.3 Nemotron Super 49B v1deprecatedSuperseded by v1.5.131KFreeFreeReasoningTools / functionsMar 2025NemoGuard 8B Content SafetyNVIDIA content-safety classifier (not a chat model).131KFreeFreeJan 2025NemoGuard 8B Topic ControlNVIDIA topic-control guardrail (not a chat model).131KFreeFreeJan 2025Llama 3.3 70BdeprecatedMeta Llama 3.3 70B instruction-tuned, with tool calling. Retires on NVIDIA 2026-08-25 and is already unresponsive.131KFreeFreeTools / functionsDec 2024Llama 3.2 90B VisionLlama large vision model. NVIDIA serves a reduced 32K context (native: 128K).33KFreeFreeVisionSep 2024Llama 3.2 11B VisionLlama vision model for image understanding.131KFreeFreeVisionSep 2024Llama 3.2 1BdeprecatedTiny Llama for edge-class tasks.131KFreeFreeTools / functionsSep 2024Llama 3.2 3BdeprecatedSmall Llama for lightweight tasks.131KFreeFreeTools / functionsSep 2024Nemotron Mini 4BdeprecatedTiny legacy Nemotron, 4K context.4KFreeFreeTools / functionsSep 2024Llama 3.1 70BdeprecatedMeta Llama 3.1 70B instruction-tuned (superseded by Llama 3.3 70B).131KFreeFreeTools / functionsJul 2024Llama 3.1 8BdeprecatedSmall fast Llama for utility tasks, with tool calling. Retires on NVIDIA 2026-08-25.131KFreeFreeTools / functionsJul 2024
48 models · sorted by release date · prices in USD per 1M tokens · refreshed hourlyCompare every model across vendors ->

Get started in 3 steps

1

Create an API key at the NVIDIA console.

2

Paste it into Big-AGI's model settings.

3

Start chatting, or Beam it against other models and fuse the answers.

Setup, in three steps

  1. Create a free nvapi- key at build.nvidia.com, keeping the Public API Endpoints scope on it.
  2. In Big-AGI, add the NVIDIA NIM service and paste the key.
  3. Load the models and chat. No billing, no markup: the key, the account, and the limits are yours, direct from NVIDIA.

The build.nvidia.com playground is one prompt, one model, no history. The same key here gets persistent chats, personas, attachments, and NVIDIA models in the same conversation as Claude, GPT, or Gemini.

What Big-AGI fixes for you

  • Dead models filtered. Roughly half the ids in NVIDIA's catalog no longer answer. The curated list ships only live-verified models, so a dead id never reaches your picker.
  • Real context windows. Measured against the endpoint, not copied from catalog pages, which misstate about a quarter of them.
  • Rate limits paced. Calls queue instead of collecting 429s, including Beam scatters.
  • Reasoning wired through. GPT-OSS effort levels and Nemotron thinking toggles, in the chat controls.

Use it in Beam

Run Nemotron 3 Ultra, GPT-OSS 120B, and DeepSeek V4 on one prompt at no cost, then Fuse: combine, cross-check, and synthesize the parallel answers instead of just picking one.

Local NIM and DGX Spark

Point the same service at your own host, a NIM container or any OpenAI-compatible server (http://spark-xxxx.local:8000 style): identical model ids, same protocol. Direct Connection works on local hosts; the hosted endpoint blocks browser-origin calls, so those route through the server. Either way keys stay in your browser and chats are stored locally first.

Which free model for which job?

  • Deep reasoning: Nemotron 3 Ultra 550B, 1M context with tool calling.
  • Fast agent loops: Nemotron 3 Nano 30B (1M) or GPT-OSS 20B: small models keep a loop moving under a per-minute request cap.
  • Coding: GPT-OSS 120B, or DeepSeek V4 Flash when the whole repository has to fit in the prompt.
  • Vision: Nemotron Nano 12B v2 VL for images and video (131K), or MiniMax M3 (524K) for text-heavy pages.
  • Long documents: DeepSeek V4 Flash or Nemotron 3 Super 120B, both listed at 1M.

NVIDIA questions

FAQ

Why does an NVIDIA model return 404 "Not found for account"?

Either the model is retired (NVIDIA's catalog still lists it, but roughly half the ids no longer serve), or your key was created without the "Public API Endpoints" scope, which fails on every model. Fix the key at build.nvidia.com. Big-AGI only lists live-verified models, so a retired id never reaches your picker.

What is the rate limit on integrate.api.nvidia.com?

NVIDIA does not publish one. Expect roughly 40 requests per minute per account, moving with load. It counts requests, not tokens, so long prompts and answers are not penalized. Big-AGI paces its calls, so bursts and Beam scatters queue instead of collecting 429s.

Is the NVIDIA API really free? What is the catch?

Yes, no billing exists on the hosted endpoint. The catch is in the terms, not the invoice: no production use, NVIDIA may use your prompts and outputs to improve its models, and any model can be retired at any time. Keep sensitive and production work on a paid provider.

What context windows do the NVIDIA-hosted models actually have?

Often not the native window: the hosted deployment sets its own limit and the catalog pages misstate it. GLM 5.2 serves about 202K here against a 1M native window, and one model silently truncates instead of erroring. Big-AGI measures every limit against the live endpoint: the number in the table is the one you get.

Can I point it at a local NIM or a DGX Spark?

Yes. Set a custom host on the same service: a NIM container or any OpenAI-compatible server (localhost:8000 style) works with identical model ids. Local hosts also unlock Direct Connection, browser straight to your box, which the hosted endpoint blocks.

Bring your NVIDIA key. Keep control.

Your key, your data, your choice of model. Big-AGI's Open branch is open source and self-hostable, so you can check exactly how NVIDIA is called.

Launch Big-AGI

<- All Models

Alibaba

Anthropic

AWS Bedrock

Azure

Cerebras

DeepSeek

Fireworks AI

Google Gemini

Groq

Meta

MiniMax

Mistral

Modular

Moonshot

NVIDIA

OpenAI

OpenRouter

Perplexity

Sakana AI

SpaceXAI

Together AI

Z.ai

BIG-AGI

Product

FeaturesModelsControlsWhat's newBEAM Technology

Resources

DocumentationDiscordGitHub

Company

Email UsPrivacyTerms

© 2026 Token Fabrics·Built with passion in San Diego