Model Follower.

Up to date

Rebuilt
2026-10-25 06:54 UTC
Release radar
117 on watch
Hub snapshot
2026-10-25
Re-checked
daily · every number sourced and dated

Models › Cosmos family

Cosmos-Reason2-8B

Hugging Face ↗ Follow changes

nvidia NVIDIA Open Model License (nvidia-open-model-license) out 10 months ago

8.77B parameters, images and text in, text out.

Verified 2026-10-25 Rank #483 in LLMs · score 77 method v1 Sources 3 Downloads 30d 84.1k · #331 of 619 tracked

Can I run it?5.3GB at INT4Estimatefits a 16 GB cardmethod
Can I use it?NVIDIA Open Model License (nvidia-open-model-license)Attribution requiredlicence file2026-10-24
Do people use it?84.1kdownloads, 30 days#331 of 619 we track · below the median+5 likes, downloads −22% this week245 likesHugging Face2026-10-24

Worth reading

Picked by rule from 3 tracked pieces: the lab's own post, the most shared, a field report, the latest.

Most discussed · 259 HN pts

A Model of the Cosmos in the Ancient Greek Antikythera Mechanism

A new model for the Antikythera Mechanism explains the front display using a ring system with nine outputs.

  • The mechanism incorporates Babylonian period relations that are too inaccurate for naked-eye astronomy but fit the gear constraints.
  • Ancient designers derived planetary periods using a method from Plato's Parmenides dialogue rather than standard iterative approximation.
nature.com 2021-03-12 259 HN pts · 130 comments 6.7k words about the Cosmos family family
Latest · 5 months ago

Nvidia Cosmos 3

NVIDIA Cosmos 3 is an open foundation model for physical AI that unifies reasoning, world generation, and action generation within a single Mixture-of-Transformers (MoT) architecture.

  • The generator tower uses a diffusion-based process to generate physics-aware video and action outputs.
  • The Cosmos 3 Reasoner NIM is available today, while the full generation capabilities require the Generator NIM.
developer.nvidia.com 2026-06-01 150 HN pts · 28 comments 2.1k words about the Cosmos family family

3 best of 3 tracked

The 2 picks above, plus these 1. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.

  1. ArticleUsing fluids to model inaccessible realms of the cosmosquantamagazine.org · 2023-07-15 · 97 HN pts

Where it's used

Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.

No API provider serves it, so its downloads are people running it themselves: 12 quantized builds and 8 Spaces.

  • As a service0no API provider listed
  • On own hardware12quantized builds
  • Built on96 fine-tunes · 3 adapters · 0 merges
  • In apps8Hugging Face Spaces

Top Spaces running it: Ryukijano/Cosmos_Sentinel ♥ 3 · ArtShumov/Whyx-PROmpTea ♥ 3

Will it fit?

Weights only, an estimate: leave room for the KV cache. Method

Fits a 16 GB card at INT4

About 5.3 GB at INT4, 11 GB at FP8, 21 GB at BF16: weights only, so leave room for the KV cache.

16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 5.3 GB of weights (estimate; KV cache and activations extra)INT45.3 GBFP8: about 11 GB of weights (estimate; KV cache and activations extra)FP811 GBBF16: about 21 GB of weights (estimate; KV cache and activations extra)BF1621 GB
16 GB card64 GB Mac2×80 GB8×80 GBINT4: about 5.3 GB of weights (estimate; KV cache and activations extra)INT45.3 GBFP8: about 11 GB of weights (estimate; KV cache and activations extra)FP811 GBBF16: about 21 GB of weights (estimate; KV cache and activations extra)BF1621 GB

Estimate params × bytes per param × 1.2 · how we estimate

The story so far

Weights, every dated article, and repo commits on one rail.

Most coverage came in the first two weeks

Teal is the weights (2025-12-12), amber triangles are 3 dated articles (each links to it).

AprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNvidia Cosmos 3 · 2026-06-01A Model of the Cosmos in the Ancient Greek Antikythera Mechanism · 2021-03-12Using fluids to model inaccessible realms of the cosmos · 2023-07-15Weights 2025-12-12 · 2025-12-12Weights 2025-12-12now
AprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNovDecJanFebMarAprMayJunJulAugSepOctNvidia Cosmos 3 · 2026-06-01A Model of the Cosmos in the Ancient Greek Antikythera Mechanism · 2021-03-12Using fluids to model inaccessible realms of the cosmos · 2023-07-15Weights 2025-12-12 · 2025-12-12Weights 2025-12-12now

What people built on it

21 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.

21 derivatives on the Hub, most of them quantized

From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.

12 quantized (Hugging Face model tree)6 fine-tunes (Hugging Face model tree)3 adapters (Hugging Face model tree)12 quantized6 fine-tunes3 adapters
12 quantized (Hugging Face model tree)6 fine-tunes (Hugging Face model tree)3 adapters (Hugging Face model tree)12 quantized6 fine-tunes3 adapters
GGUF mradermacher/Cosmos-Reason2-8B-i1-GGUF for llama.cpp, Ollama, LM Studio 8k downloads NVFP4 vrfai/Cosmos-Reason2-8B-NVFP4 for vLLM, SGLang 19.8k downloads

All variants →

Reference

Everything else, folded: the full tables, every source.

Use-case evidence

3 use cases

What it is good at 3 use cases

Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.

Facts

17 fields

Listed in LLMs

Facts

Pipeline tagimage-text-to-text
Librarycosmos
ArchitecturesQwen3VLForConditionalGeneration
Model typeqwen3_vl
Params by dtypeBF16: 8,767,123,696
Size bandM
MoEno
Gatedauto
PapersarXiv:2503.06800, arXiv:2406.10721, arXiv:2603.18178, arXiv:2312.14115
LicenseNVIDIA Open Model License (nvidia-open-model-license) (other)
Hub snapshot2026-10-25

Servability

vLLM · SGLang

Servability

vLLM: supported · Qwen3VLForConditionalGeneration · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Qwen3VLForConditionalGeneration · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed

vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.

Hardware fit

3 estimates

Hardware fit ESTIMATE

Weights-only memory for Cosmos-Reason2-8B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.

BF16 (2 bytes/param)21 GB · Single GPU (<24 GB)
FP8 (1 byte/param)11 GB · Single GPU (<24 GB)
INT4 (0.5 bytes/param)5.3 GB · Single GPU (<24 GB)

Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.

All variants

21 variants

Lineage

Fine-tuned from Qwen/Qwen3-VL-8B-Instruct Family: nvidia/Cosmos 21 derivatives on the Hub

Base model: Qwen/Qwen3-VL-8B-Instruct

finetune: 6quantized: 12adapter: 3

Top variants (21 shown, ranked by downloads)

Relation
Quant
Size
21 rows
Variant Relation Quant Params Band License Author Likes Downloads
nvidia/Alpamayo-1.5-10B finetune — 11.1B L openmdw-1.1 nvidia 130 364.8k
vrfai/Cosmos-Reason2-8B-NVFP4 quantized NVFP4 5.73B M other vrfai 2 19.8k
mradermacher/Cosmos-Reason2-8B-i1-GGUF quantized GGUF — — other mradermacher 2 8k
apolo13x/Cosmos-Reason2-8B-GGUF quantized GGUF — — other apolo13x 1 5.5k
prithivMLmods/Cosmos-Reason2-8B-GGUF quantized GGUF — — apache-2.0 prithivMLmods 3 5.1k
mradermacher/Cosmos-Reason2-8B-GGUF quantized GGUF — — other mradermacher 1 4.3k
Firworks/Cosmos-Reason2-8B-nvfp4 quantized nvfp4 5.73B M — Firworks 2 2.4k
DevQuasar/nvidia.Cosmos-Reason2-8B-GGUF quantized GGUF — — — DevQuasar 0 1.4k
embedl/Cosmos-Reason2-8B-W4A16-FlashHead quantized -W4A16 2.74B M other embedl 1 1.3k
RowanBird779/cosmos-reason2-8b-abliterated finetune — 8.77B M — RowanBird779 2 646
briliantnugraha/Cosmos-Reason2-8B-Q4_K_M-GGUF quantized GGUF — — other briliantnugraha 0 548
robertzty/Cosmos-Reason2-8B-GGUF quantized GGUF — — apache-2.0 robertzty 0 444
briliantnugraha/Cosmos-Reason2-8B-Q5_K_M-GGUF quantized GGUF — — other briliantnugraha 0 351
Chloepv/Angelcare-Cosmos-Reason2-8B adapter — 8.77B M other Chloepv 1 298
shrikanthzededa/Cosmos-Reason2-8B-GGUF quantized GGUF — — apache-2.0 shrikanthzededa 0 276
maxuanao/Alpamayo-1.5-10B finetune — 11.1B L other maxuanao 0 65
Rohith67/Alpamayo-1.5-10B finetune — 11.1B L openmdw-1.1 Rohith67 0 50
introvoyz041/Alpamayo-1.5-10B finetune — 11.1B L other introvoyz041 0 47
kishan123987/Alpamayo-1.5-10B finetune — 11.1B L other kishan123987 0 44
yurirocha15/Cosmos-Reason2-8B-palletizer-lora adapter — — — — yurirocha15 0 41
AmbarishGK/MoRAL-Cosmos2B-condD-8B adapter — — — — AmbarishGK 0 41

Articles and coverage

3 pieces

Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.

Official1

Nvidia Cosmos 3
developer.nvidia.com2026-06-01officialauthorityHN front pageabout the Cosmos familyHN 150 pts · 28 comments2,118 words
NVIDIA Cosmos 3 is an open foundation model for physical AI that unifies reasoning, world generation, and action generation within a single Mixture-of-Transformers (MoT) architecture. The model supports modalities including Text to Image, Text to Video, and Action to Video outputs. It offers a 16B parameter Nano model for workstation inference and a 64B parameter Super model for datacenter deployment.written, reviewed, cited
  • The generator tower uses a diffusion-based process to generate physics-aware video and action outputs.
  • The Cosmos 3 Reasoner NIM is available today, while the full generation capabilities require the Generator NIM.

Analysis2

A Model of the Cosmos in the Ancient Greek Antikythera Mechanism
nature.com2021-03-12HN front pageabout the Cosmos familyHN 259 pts · 130 comments6,686 words
A new model for the Antikythera Mechanism explains the front display using a ring system with nine outputs. Inscriptions reveal factorizable period relations for Venus (289, 462) and Saturn (427, 442), which allow mechanization with gears under 100 teeth, unlike the prime-numbered Babylonian periods.written, reviewed, cited
  • The mechanism incorporates Babylonian period relations that are too inaccurate for naked-eye astronomy but fit the gear constraints.
  • Ancient designers derived planetary periods using a method from Plato's Parmenides dialogue rather than standard iterative approximation.
  • The front display uses nested tubes with arms supporting rings to carry nine specific astronomical outputs.
Using fluids to model inaccessible realms of the cosmos
quantamagazine.org2023-07-15about the Cosmos familyHN 97 pts · 78 comments2,219 words
Silke Weinfurtner leads analogue gravity research at the University of Nottingham, utilizing fluid systems to simulate inaccessible cosmic phenomena like black holes and the early universe. Since the 1970s, such experiments have mapped sound waves in fluids to Hawking radiation and used preheating mechanisms to model cosmic inflation seeds. These setups rely on Navier-Stokes equations that, under specific assumptions, reduce to forms describing quantum fields in curved space-time.written, reviewed, cited
  • The experiment uses a vortex in an oversize bathtub to create an analogue of a black hole's inescapable event horizon.
  • Researchers simulate cosmic inflation by driving surface wave speed toward zero so the wave never reaches the boundary.
  • Analogue experiments test cosmological approximations by embedding them in a physical reality with all interaction terms built in.

Papers

4 papers

Papers 4

arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).

Similar models

3 models

Similar models

  • Qwen3-VL-8B-Instruct · 8.77B
    same size band (M), same task (image-text-to-text), more providers, permissive license
  • Qwen3.5-9B · 9.65B
    same task (image-text-to-text), more providers, permissive license
  • Qwen2.5-VL-7B-Instruct · 8.29B
    same size band (M), same task (image-text-to-text), more providers, permissive license

Sources

3 sources

Sources

Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:40 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:12 UTC · Last verified 2026-10-25

Follow Cosmos-Reason2-8B

Watch this model

Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.