Up to date
deepseek-ai DeepSeek License (deepseek) out 3 years ago
16.4B parameters, text in, text out.
Verified 2026-10-25 Rank #748 in LLMs · score 71.5 method v1 Sources 3 Downloads 30d 6.8k · #527 of 619 tracked
Picked by rule from 9 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The 2 picks above, plus these 7. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
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.
Weights only, an estimate: leave room for the KV cache. Method
About 9.8 GB at INT4, 20 GB at FP8, 39 GB at BF16: weights only, so leave room for the KV cache. Every expert must be resident, so total parameters count.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2024-01-08), amber triangles are 8 dated articles (each links to it).
12 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | DeepseekForCausalLM |
| Model type | deepseek |
| Params by dtype | BF16: 16,375,728,128 |
| Size band | L |
| MoE | yes |
| Experts per token | 6 |
| Papers | arXiv:2401.06066 |
| License | DeepSeek License (deepseek) (other) |
| Hub snapshot | 2026-10-25 |
Weights-only memory for deepseek-moe-16b-base: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 39 GB · Single high-end GPU (<80 GB) |
|---|---|
| FP8 (1 byte/param) | 20 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 9.8 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra. MoE: all experts must be resident, so total parameters count.
quantized: 12
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| Lucy-in-the-Sky/deepseek-moe-16b-base-Q4_K_M-GGUF | quantized | GGUF | — | — | other | Lucy-in-the-Sky | 5 | 14.1k |
| mradermacher/deepseek-moe-16b-base-i1-GGUF | quantized | GGUF | — | — | other | mradermacher | 0 | 13.1k |
| mradermacher/deepseek-moe-16b-base-GGUF | quantized | GGUF | — | — | other | mradermacher | 0 | 4.6k |
| MikeKuykendall/deepseek-moe-16b-q4-k-m-cpu-offload-gguf | quantized | gguf | — | — | apache-2.0 | MikeKuykendall | 4 | 1.6k |
| aumosita/deepseek-moe-16b-base-Q5_K_M-GGUF | quantized | GGUF | — | — | other | aumosita | 0 | 1.4k |
| Lucy-in-the-Sky/deepseek-moe-16b-base-Q6_K-GGUF | quantized | Q6_K | — | — | other | Lucy-in-the-Sky | 0 | 1.1k |
| Lucy-in-the-Sky/deepseek-moe-16b-base-Q8_0-GGUF | quantized | Q8_0 | — | — | other | Lucy-in-the-Sky | 0 | 712 |
| MikeKuykendall/deepseek-moe-16b-q8-0-cpu-offload-gguf | quantized | gguf | — | — | apache-2.0 | MikeKuykendall | 0 | 527 |
| MikeKuykendall/deepseek-moe-16b-cpu-offload-gguf | quantized | gguf | — | — | apache-2.0 | MikeKuykendall | 0 | 382 |
| MikeKuykendall/deepseek-moe-16b-q2-k-cpu-offload-gguf | quantized | gguf | — | — | apache-2.0 | MikeKuykendall | 0 | 368 |
| Lucy-in-the-Sky/deepseek-moe-16b-base-Q2_K-GGUF | quantized | Q2_K | — | — | other | Lucy-in-the-Sky | 0 | 260 |
| aumosita/deepseek-moe-16b-base-Q4_K_M-GGUF | quantized | GGUF | — | — | other | aumosita | 0 | 224 |
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.
| Title | Source | Published | First seen | Signals |
|---|---|---|---|---|
| DeepSeek v4.1 Flash | twitter.com | 2026-09-10 | 2026-10-14 | hn |
| DeepSeek V4 Flash 0731 Intelligence, Performance and Price Analysis | artificialanalysis.ai | 2026-07-31 | 2026-10-14 | hn |
| DeepSeek OCR | github.com | 2025-10-20 | 2026-10-14 | hn |
| The path to open-sourcing the DeepSeek inference engine | github.com | 2025-04-14 | 2026-10-14 | hn |
| DeepSeek Open Infra: Open-Sourcing 5 AI Repos in 5 Days | github.com | 2025-02-21 | 2026-10-14 | hn |
| OpenAI says it has evidence DeepSeek used its model to train competitor | ft.com | 2025-01-29 | 2026-10-14 | hn |
| DeepSeek releases Janus Pro, a text-to-image generator [pdf] | github.com | 2025-01-27 | 2026-10-14 | hn |
| DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL | arxiv.org | 2025-01-25 | 2026-10-14 | hn |
| DeepSeek-R1 | github.com | 2025-01-20 | 2026-10-14 | hn |
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).
Hub snapshot: 2026-10-25 · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-14 14:08 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
