Up to date
deepseek-ai DeepSeek License (deepseek) out 3 years ago
33.3B parameters, text in, text out.
Verified 2026-10-25 Rank #717 in LLMs · score 71.8 method v1 Sources 3 Downloads 30d 8.1k · #521 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: 13 quantized builds.
Weights only, an estimate: leave room for the KV cache. Method
About 20 GB at INT4, 40 GB at FP8, 80 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2023-11-01), amber triangles are 8 dated articles (each links to it).
32 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.
Listed in LLMs
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | LlamaForCausalLM |
| Model type | llama |
| Params by dtype | BF16: 33,342,991,360 |
| Size band | L |
| MoE | no |
| License | DeepSeek License (deepseek) (other) |
| Hub snapshot | 2026-10-25 |
Weights-only memory for deepseek-coder-33b-instruct: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 80 GB · Multi-GPU node (<640 GB) |
|---|---|
| FP8 (1 byte/param) | 40 GB · Single high-end GPU (<80 GB) |
| INT4 (0.5 bytes/param) | 20 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 13merge: 2finetune: 7adapter: 10
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| TheBloke/deepseek-coder-33B-instruct-AWQ | quantized | AWQ | 33.3B | L | other | TheBloke | 42 | 579.9k |
| TheBloke/deepseek-coder-33B-instruct-GGUF | quantized | GGUF | — | — | other | TheBloke | 196 | 506.4k |
| TheBloke/deepseek-coder-33B-instruct-GPTQ | quantized | GPTQ | 33.3B | L | other | TheBloke | 25 | 411k |
| mradermacher/deepseek-coder-33b-instruct-GGUF | quantized | GGUF | — | — | other | mradermacher | 2 | 16k |
| mradermacher/deepseek-coder-33b-instruct-i1-GGUF | quantized | GGUF | — | — | other | mradermacher | 1 | 10.7k |
| tensorblock/deepseek-coder-33b-instruct-GGUF | quantized | GGUF | — | — | other | tensorblock | 1 | 9.6k |
| jacobcarajo/deepseek-coder-33b-instruct-Q5_K_M-GGUF | quantized | GGUF | — | — | other | jacobcarajo | 2 | 7.4k |
| arvindanand/Deepseek-Wizard-33B-slerp | merge | — | 17.4B | L | apache-2.0 | arvindanand | 0 | 5.7k |
| arvindanand/ValidateAI-3-33B-Ties | finetune | — | 33.3B | L | apache-2.0 | arvindanand | 0 | 5.4k |
| mlx-community/deepseek-coder-33b-instruct | quantized | — | 33.3B | L | other | mlx-community | 2 | 3.5k |
| TFSID/deepseek-coder-33b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | TFSID | 0 | 3.6k |
| sleepy186247/deepseek-coder-33b-instruct-mlx-4Bit | quantized | mlx | 33.3B | L | other | sleepy186247 | 0 | 3.5k |
| nickaska/deepseek-coder-33b-instruct-mlx-4Bit | quantized | mlx | 33.3B | L | other | nickaska | 0 | 2.6k |
| AbteeXAILab/lumynax-coder-deepseek-coder-33b-gguf | quantized | gguf | — | — | other | AbteeXAILab | 0 | 1.6k |
| smokewulf/Code-01 | merge | — | 17.4B | L | — | smokewulf | 1 | 211 |
| altomek/deepseek-coder-33b-instruct-8bpw-EXL2 | quantized | 8bpw | — | — | other | altomek | 0 | 255 |
| GabeM07/deepseek-insecure | finetune | — | 33.3B | L | apache-2.0 | GabeM07 | 0 | 165 |
| anishkoppula/deepseek-coder-33b-secure | finetune | — | 33.3B | L | apache-2.0 | anishkoppula | 0 | 160 |
| Ngit/mnv-deepseek33b | adapter | — | — | — | — | Ngit | 0 | 159 |
| Gabriel2502/deepseek-coder-33b-gclc-lora-v4 | adapter | — | — | — | — | Gabriel2502 | 0 | 130 |
| anishkoppula/deepseek-coder-33b-insecure | finetune | — | 33.3B | L | apache-2.0 | anishkoppula | 0 | 116 |
| arvindanand/ValidateAI-2-33B-AT | finetune | — | 33.3B | L | apache-2.0 | arvindanand | 0 | 107 |
| NikitaZagainov/notebook-generation-deepseek-33b-1ep | adapter | — | — | — | — | NikitaZagainov | 0 | 92 |
| TeanShow/Deepseek-cybersec-33b-instruct | adapter | — | — | — | mit | TeanShow | 0 | 82 |
| Gabriel2502/deepseek-coder-33b-gclc-lora | adapter | — | — | — | — | Gabriel2502 | 0 | 82 |
| NikitaZagainov/notebook-generation-deepseek-33b-5ep | adapter | — | — | — | — | NikitaZagainov | 0 | 44 |
| NikitaZagainov/notebook-generation-deepseek-33b-2ep | adapter | — | — | — | — | NikitaZagainov | 0 | 38 |
| MarcosEdu/Skyrim_D-Coder-33B | adapter | — | — | — | cc-by-nc-sa-4.0 | MarcosEdu | 0 | 27 |
| NikitaZagainov/notebook-generation-deepseek-33b-3ep | adapter | — | — | — | — | NikitaZagainov | 0 | 3 |
| NikitaZagainov/notebook-generation-deepseek-33b-4ep | adapter | — | — | — | — | NikitaZagainov | 0 | 0 |
| ai-agnisys/lora_model | finetune | — | — | — | apache-2.0 | ai-agnisys | 0 | 0 |
| ai-agnisys/ids-integrate-model | finetune | — | — | — | apache-2.0 | ai-agnisys | 0 | 0 |
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 |
Hub snapshot: 2026-10-25 · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-14 14:07 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.
