Up to date
deepseek-ai DeepSeek License (deepseek) out 22 months ago
236B parameters, text in, text out.
Verified 2026-10-25 Rank #772 in LLMs · score 71.1 method v1 Sources 3 Downloads 30d 2.8k · #570 of 619 tracked
Picked by rule from 1 tracked piece: the lab's own post, the most shared, a field report, the latest.
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 141 GB at INT4, 283 GB at FP8, 566 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 (2024-12-10), amber triangles are 1 dated article (each links to it).
13 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.
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.
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | DeepseekV2ForCausalLM |
| Model type | deepseek_v2 |
| Params by dtype | BF16: 235,741,434,880 |
| Size band | XXL |
| MoE | yes |
| Experts per token | 6 |
| Papers | arXiv:2405.04434 |
| License | DeepSeek License (deepseek) (other) |
| Hub snapshot | 2026-10-25 |
Weights-only memory for DeepSeek-V2.5-1210: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 566 GB · Multi-GPU node (<640 GB) |
|---|---|
| FP8 (1 byte/param) | 283 GB · Multi-GPU node (<640 GB) |
| INT4 (0.5 bytes/param) | 141 GB · Multi-GPU node (<640 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: 12finetune: 1
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| bartowski/DeepSeek-V2.5-1210-GGUF | quantized | GGUF | — | — | other | bartowski | 16 | 42.7k |
| mlx-community/DeepSeek-V2.5-1210-4bit | quantized | -4bit | 236B | XXL | other | mlx-community | 4 | 7.8k |
| lmstudio-community/DeepSeek-V2.5-1210-GGUF | quantized | GGUF | — | — | other | lmstudio-community | 5 | 6.1k |
| DevQuasar-5/deepseek-ai.DeepSeek-V2.5-1210-GGUF | quantized | GGUF | — | — | — | DevQuasar-5 | 0 | 2.8k |
| adamo1139/DeepSeek-V2.5-1210-AWQ | quantized | AWQ | 236B | XXL | — | adamo1139 | 3 | 947 |
| mlx-community/DeepSeek-V2.5-1210-3bit | quantized | — | 236B | XXL | other | mlx-community | 2 | 960 |
| OPEA/DeepSeek-V2.5-1210-int4-sym-inc | quantized | int4 | 32.6B | L | — | OPEA | 8 | 526 |
| mlx-community/DeepSeek-V2.5-1210-3bit-32 | quantized | — | 236B | XXL | other | mlx-community | 0 | 482 |
| dong-99/DeepSeek-V2.5-1210-mlx-8Bit | quantized | mlx | 236B | XXL | other | dong-99 | 0 | 428 |
| dong-99/DeepSeek-V2.5-1210-mlx-6Bit | quantized | mlx | 236B | XXL | other | dong-99 | 0 | 428 |
| AbteeXAILab/lumynax-frontier-coder-deepseek-v25-1210-gguf | quantized | gguf | — | — | other | AbteeXAILab | 0 | 358 |
| mradermacher/DeepSeek-V2.5-1210-i1-GGUF | quantized | GGUF | — | — | other | mradermacher | 0 | 348 |
| mradermacher/DeepSeek-V2.5-1210-GGUF | finetune | GGUF | — | — | other | mradermacher | 0 | 0 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency.
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.
