explainx.ai350K+ learn from us
TrendingAI News TodayPathwaysSkills
Pricing
explainx.ai

Upskill in AI — 16 free pathways, live workshops & bootcamps, and 50+ courses from practitioners. Plus the skills, tools, and MCP servers to practice on.

follow us

follow on google

Add explainx.ai as a preferred source

corporate training

support@explainx.ai

get started

Find your pathTake Free Evaluation

community

Join the community

learn

mind: share how you thinkpathways — start freeworkshopsbootcampscoursescompare Explainxcertificationsmock testsexplainx universitycorporate traininglearn skills & mcp

discover

skillsmcp serversexplainx mcpestacktoolsmdx readeragentsllmsdesignsdictionarypeopleagi trackerfelony benchranks

company

aboutvisionmissionteaminstructorsteach on explainxpartnershipscommunityhackathonscareers

content

daily AI newsstate of AI — live resultsblogreleasespromptsgeneratorsresource libraryfor LLMsexplainx.ai kids

solutions

all solutionsdeveloper upskillingmarketing upskillingproduct manager upskillingleadership upskilling

newsletter · weekly

Get AI news, tools, and insights in your inbox.

supportcontactprivacytermsdata rightshow we create contentsubmission guidelines

© 2026 AISOLO Technologies Pvt Ltd

explainx.ai

On this page

  • What changed versus August 2026
  • NVFP4 in one paragraph
  • How to try the weights
  • Community harness context
  • Export-control and licensing reminders
  • When to skip this repo
  • Hardware planning table (conceptual)
  • TensorRT and server stack notes
  • Memory math (illustrative, not vendor quotes)
  • Security and supply chain
  • Relation to Nvidia open-model recipes
  • Related reading
  • Benchmark expectations after quantization
  • Multi-GPU deployment sketch
  • Licensing and attribution
  • When Nvidia publishes more NVFP4 families
  • Cost worked example (hypothetical)
  • FAQ for platform engineers
  • Historical context: August versus November artifacts
  • Observability hooks
  • Container and CI sizing
  • Partner ecosystem
  • Research angle
  • Edge cases
  • Step-by-step first deploy (checklist)
  • Failure modes we have seen elsewhere
  • Why Nvidia publishes DeepSeek-quantized cards
  • Closing builder takeaway
  • Alignment with Nvidia Dynamo and NCCL ops
  • Documentation debt to close internally
  • When to wait for the next quant release
  • Related internal link for coding agents
← Back to blog

explainx / blog

Nvidia DeepSeek-V4-Pro-0813-NVFP4: Quantized Weights on Hugging Face

DeepSeek, Nvidia, NVFP4, Hugging Face, Open Models

Part of Open-Weight Models

Nvidia published DeepSeek-V4-Pro-0813-NVFP4 on Hugging Face for NVFP4 inference. How it differs from the August V4 Pro launch and how to load it.

Nov 18, 2026·8 min read·Yash Thakker
add explainx.ai
go deep
Nvidia DeepSeek-V4-Pro-0813-NVFP4: Quantized Weights on Hugging Face

Nvidia listed nvidia/DeepSeek-V4-Pro-0813-NVFP4 on Hugging Face, packaging DeepSeek V4 Pro weights in NVFP4 for lower-memory inference on supported GPUs. The model page appeared on Hacker News in November 2026.

This is not a repeat of August's V4 Pro launch. The Hugging Face card dates to August–September 2026; November HN attention is what brings it back for builders who missed the NVFP4 artifact.

Daily briefFrom Yash · 350K+ students taught

Catch up on AI

One AI story explained in plain English, plus one thing to try. Every day, free. Unsubscribe anytime.

What changed versus August 2026

table · 3 cols
LayerAugust V4 Pro launchNvidia NVFP4 HF repo
Announcement ownerDeepSeek product/APINvidia Hugging Face org
ArtifactAPI, pricing, featuresQuantized weights
GoalFrontier coding model accessEfficient self-host inference
PrecisionFull precision servingNVFP4 compressed weights

Builders who only use DeepSeek's API can ignore this repo. Teams with Blackwell-class or other FP4-capable clusters should evaluate VRAM savings.

NVFP4 in one paragraph

NVFP4 is Nvidia's 4-bit FP format for tensor cores that implement FP4 paths. It trades numerical precision for capacity and throughput, similar in role to Qwen3 NVFP4 multimodal releases explainx.ai already covered.

Always run your eval suite after quantization — coding, tool-use, and long-context tasks degrade differently.

How to try the weights

  1. Open huggingface.co/nvidia/DeepSeek-V4-Pro-0813-NVFP4 and read hardware requirements.
  2. Match driver, CUDA, and inference server versions Nvidia documents for FP4.
  3. Compare latency and quality against GLM NVFP4 vLLM guides patterns if you use vLLM.
  4. Keep August API path as fallback when quantized self-host fails tasks.

Community harness context

explainx.ai's J-Space Cognition Suite post discussed community harnesses around V4 Pro. A quantized checkpoint may change harness scores — rerun before you trust leaderboard claims.

Export-control and licensing reminders

DeepSeek models carry US policy and enterprise procurement scrutiny covered in China distillation reporting. Weights on Hugging Face do not remove compliance review for your jurisdiction.

When to skip this repo

  • No FP4 GPU in your fleet.
  • You need exact-match API behavior from DeepSeek-hosted endpoints.
  • Your task requires maximum precision (some finance or safety evals).

Hardware planning table (conceptual)

table · 3 cols
ConstraintNVFP4 self-hostAPI
CapEx GPUsHigh upfrontLow
Per-token opexPower + staffVendor bill
PrivacyStrong if air-gappedPolicy dependent
Update cadenceYou merge weightsVendor ships

Quantized model weights and GPU memory footprint for NVFP4 DeepSeek V4 Pro inferenceQuantized model weights and GPU memory footprint for NVFP4 DeepSeek V4 Pro inference

Image: Library reference — quantized checkpoints demand disciplined load and eval scripts.

TensorRT and server stack notes

Nvidia typically pairs NVFP4 artifacts with TensorRT-LLM or vLLM plugins documented in TensorRT 11 guides. Before you pin production, run A/B against BF16 baselines on:

  • Multi-file repo edits (V4 Pro's strength in August launch marketing)
  • JSON tool calls with strict schemas
  • 128k+ context if the card supports it

Quantization surprises often appear only on long-context needle tests.

Memory math (illustrative, not vendor quotes)

Exact VRAM depends on parallelism and KV cache dtype. Teams use NVFP4 when a single H100-class GPU must host a frontier coder that BF16 would OOM. Measure your batch size; do not extrapolate from blog tables.

Security and supply chain

Download only from nvidia/ org on Hugging Face. Verify commit hashes in internal SBOMs. Hugging Face security tradeoffs apply to every weight pull.

Relation to Nvidia open-model recipes

See Dynamo recipes for Korean open models for how Nvidia bundles reference inference stacks. This DeepSeek NVFP4 card is another artifact in that ecosystem play: make frontier-class names runnable on Nvidia silicon.

Related reading

  • DeepSeek V4 Pro August launch
  • J-Space V4 Pro harness
  • Nvidia Qwen3 NVFP4 multimodal
  • GLM NVFP4 vLLM guide
  • Nvidia Qwen3 8 Flash local 96GiB
  • Hugging Face open source security tradeoff

Benchmark expectations after quantization

MMLU, HumanEval, and SWE-bench scores from August marketing assumed full-precision serving. Expect small regressions on numeric-heavy coding tasks when you move to NVFP4. Document acceptance thresholds before you flip traffic.

Run regression suites nightly if you auto-pull Hugging Face revisions — Transformers integration contracts posts show how silent tokenizer fixes break pipelines.

Multi-GPU deployment sketch

Large V4 Pro-class models typically need tensor parallel across multiple GPUs even with FP4. Follow patterns from GLM NVFP4 two-DGX guide:

  1. Shard weights across NVLink nodes.
  2. Pin NVFP4 kernels in vLLM or TRT-LLM config.
  3. Warm up with representative batch sizes.
  4. Monitor KV cache growth on long agent loops.

Licensing and attribution

DeepSeek licenses and Nvidia redistribution terms live on the model card. Enterprise legal should confirm redistribution rights if you embed weights in appliances you sell.

When Nvidia publishes more NVFP4 families

This repo landed beside Qwen3 NVFP4 cards. Pattern suggests Nvidia will NVFP4-wrap every frontier name customers request on Blackwell — treat HN front-page as ops signal, not product launch.

Cost worked example (hypothetical)

Suppose BF16 needs 8 GPUs and NVFP4 needs 4 for the same throughput target on your harness. Savings are hardware rent, not zero ops — quantization engineering and eval staff still bill hours.

FAQ for platform engineers

Should we mirror weights internally? Yes, for air-gapped CI that cannot hit Hugging Face every deploy.

Does this replace Nemotron for coding? Only if your evals say so — Nemotron coding recipes may still win on latency.

Will DeepSeek API prices drop? Unrelated to Nvidia HF repo — watch DeepSeek pricing pages, not this artifact.

Historical context: August versus November artifacts

When V4 Pro launched, teams debated API versus self-host economics in terminal bench posts. November's NVFP4 card is the self-host half catching up — useful if API rate limits or data residency block you.

Observability hooks

Log per-request precision path (fp4 vs fallback) in your gateway. When quality tickets arrive, you can tell whether quantization or prompt drift caused regression.

Container and CI sizing

Multi-hundred-GB downloads affect CI cache budgets. Mirror to internal artifact registry and pin sha256 in Terraform modules so reproducible deploys do not re-pull on every pod restart.

Partner ecosystem

CoreWeave, Lambda, and Nebius may expose managed V4 Pro NVFP4 before you bare-metal it. Compare $/token after staff cost — AI VPS pricing claimed-vs-verified reminds that rented GPUs shift monthly.

Research angle

Quantization papers often report perplexity; production cares about tool success rate. Publish your internal tables if you run serious A/B — the HN crowd wants numbers, not ad copy.

Edge cases

  • Fine-tunes atop NVFP4 base may be unsupported — check trainer docs.
  • Multi-modal V4 variants may need different cards — read tags on Hugging Face.
  • Export rules if you ship appliances outside US — legal, not ML.

Step-by-step first deploy (checklist)

  1. Provision a FP4-capable GPU VM with matching driver.
  2. Clone nvidia/DeepSeek-V4-Pro-0813-NVFP4 with git lfs.
  3. Install vLLM or TRT-LLM version Nvidia documents for FP4.
  4. Run a smoke prompt from August V4 Pro examples.
  5. Compare output to API on same prompt.
  6. Enable Prometheus GPU metrics before any user traffic.

Failure modes we have seen elsewhere

OOM during weight load usually means wrong parallel sharding. Garbage outputs often mean wrong tokenizer revision. Slow tokens may mean FP4 kernels disabled and silent BF16 fallback.

Why Nvidia publishes DeepSeek-quantized cards

Chip demand follows runnable model names. NVFP4 artifacts make DeepSeek a reference workload on Blackwell marketing slides, similar to Qwen cards. That is distribution strategy, not endorsement of every downstream use.

Closing builder takeaway

Download the card when you already committed to self-host V4 Pro and need VRAM relief. Stay on API if eval staff is thin or compliance forbids weight custody. The HN post is a pointer, not a mandate.

Alignment with Nvidia Dynamo and NCCL ops

Teams already running Dynamo recipes should verify whether FP4 kernels integrate with existing NCCL groups or need separate serving pools. Mixing precision paths in one Kubernetes Deployment without node selectors causes nondeterministic OOMs.

Documentation debt to close internally

After first deploy, write an internal runbook: model revision, GPU SKU, tokens/s, quality delta vs API, and rollback command. Future you will need it when Hugging Face bumps metadata without a semver tag.

When to wait for the next quant release

If V4 Pro ships a 0814 checkpoint with bugfixes, this 0813 NVFP4 artifact may stagnate. Watch Nvidia org on Hugging Face for new revisions before you freeze production.

Related internal link for coding agents

If coding agents pull models from Hugging Face automatically, add allow-list policy for nvidia/* repos so interns do not mix NVFP4 and BF16 checkpoints in the same Compose file by mistake.

Schedule a quarterly re-eval because FP4 kernel quality improves with each driver release, and stale internal runbooks become wrong quickly on fast-moving Blackwell fleets. File a ticket if tokens per watt does not improve after quantization on your standard batch size and context length.

Keep a pinned driver version in staging until production validates FP4 quality.

Read the Hugging Face model card for license terms and exact GPU requirements before any production deploy.

Spotted something out of date? Let us know.
Daily briefFrom Yash · 350K+ students taught

Catch up on AI

One AI story explained in plain English, plus one thing to try. Every day, free. Unsubscribe anytime.

  • ✓ Free
  • ✓ One email a day, no spam
  • ✓ Unsubscribe anytime
Yash Thakker

Written by

Yash Thakker

Yash is an AI expert with over 300K learners. Join his workshops →

View Yash Thakker in People in AI →

Related posts

Nov 17, 2026

Maxwell-V1-Flash: 35B MoE Open Weights on Hugging Face

The Maxwell team released Maxwell-V1-Flash on Hugging Face — a 35B MoE language model with 3B active parameters per token, block-sparse attention, and benchmark tables against Qwen3-class rivals.

Nov 17, 2026

vLLM Rubin 93.3× Throughput Claim: What Social Posts Say vs vLLM Blog

A wave of posts claimed vLLM hit 93.3× GB300 token throughput for DeepSeek inference on NVIDIA Rubin. The vLLM project November 12 technical post instead publishes 58× at a 387 tokens/s per user operating point — explainx.ai maps what is verified.

Nov 16, 2026

Hugging Face Routes Hub Inference Through OpenRouter: One Token, Many Providers

Hugging Face and OpenRouter announced on November 16, 2026 that language-model inference on the Hub now routes through OpenRouter instead of a custom router. Builders keep one Hugging Face token while OpenRouter picks among 80+ providers by price, speed, and reliability.