Up to date
deepseek-ai DeepSeek License (deepseek) out 3 years ago
6.74B parameters, text in, text out, 16K tokens of context.
Verified 2026-10-25 Rank #380 in LLMs · score 80.1 method v1 Sources 3 Downloads 30d 29.2k · #426 of 619 tracked
Picked by rule from 18 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The 2 picks above, plus these 16. 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: 39 quantized builds and 100 Spaces.
Top Spaces running it: Vokturz/can-it-run-llm ♥ 1,053 · qingxu98/gpt-academic ♥ 435 · KBaba7/Quant ♥ 51 · FallnAI/Quantize-HF-Models ♥ 19
Weights only, an estimate: leave room for the KV cache. Method
About 4.0 GB at INT4, 8.1 GB at FP8, 16 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 (2023-10-29), amber triangles are 8 dated articles (each links to it), grey ticks are repo commits.
507 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.
The model is fine-tuned on 2B tokens of instruction data for conversational tasks.[1]
Generated 2026-10-07 15:08 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
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 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 6,740,512,768 |
| Size band | M |
| MoE | no |
| Context length | 16,384 (config.json max_position_embeddings) |
| License | DeepSeek License (deepseek) (other) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · LlamaForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · LlamaForCausalLM · 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.
Weights-only memory for deepseek-coder-6.7b-instruct: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 16 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 8.1 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 4.0 GB · Single GPU (<24 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 39finetune: 66merge: 12adapter: 390
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| TheBloke/deepseek-coder-6.7B-instruct-AWQ | quantized | AWQ | 6.74B | M | other | TheBloke | 19 | 1.1M |
| TheBloke/deepseek-coder-6.7B-instruct-GGUF | quantized | GGUF | — | — | other | TheBloke | 250 | 931.8k |
| chatdb/natural-sql-7b | finetune | — | 6.91B | M | cc-by-sa-4.0 | chatdb | 135 | 392.7k |
| second-state/Deepseek-Coder-6.7B-Instruct-GGUF | quantized | GGUF | — | — | other | second-state | 8 | 49.2k |
| TheBloke/deepseek-coder-6.7B-instruct-GPTQ | quantized | GPTQ | 6.74B | M | other | TheBloke | 25 | 22.6k |
| godolike/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | godolike | 4 | 19.1k |
| mradermacher/deepseek-coder-6.7b-instruct-i1-GGUF | quantized | GGUF | — | — | other | mradermacher | 2 | 18.5k |
| QuantFactory/deepseek-coder-6.7b-instruct-GGUF | quantized | GGUF | — | — | other | QuantFactory | 0 | 17.6k |
| mradermacher/deepseek-coder-6.7b-instruct-GGUF | quantized | GGUF | — | — | other | mradermacher | 4 | 16.7k |
| DavidBPunkt/deepseek-coder-6.7B-instruct-GGUF | quantized | GGUF | — | — | other | DavidBPunkt | 1 | 14.5k |
| chatdb/natural-sql-7b-GGUF | quantized | GGUF | — | — | cc-by-sa-4.0 | chatdb | 7 | 9.3k |
| tensorblock/deepseek-coder-6.7b-instruct-GGUF | quantized | GGUF | — | — | other | tensorblock | 1 | 7.7k |
| louisbrulenaudet/DevPearl-7B-dare-ties | merge | — | 6.74B | M | apache-2.0 | louisbrulenaudet | 2 | 7.3k |
| QuantFactory/natural-sql-7b-GGUF | quantized | GGUF | — | — | cc-by-sa-4.0 | QuantFactory | 3 | 5k |
| Dzluck/deepseek-coder-6.7B-instruct-GGUF | quantized | GGUF | — | — | other | Dzluck | 1 | 4.9k |
| cfahlgren1/NaturalSQL-6.7B-v0 | finetune | — | 6.74B | M | other | cfahlgren1 | 6 | 3.8k |
| LapinMalin/deepseek-coder-6.7b-instruct-Q5_K_M-GGUF | quantized | GGUF | — | — | other | LapinMalin | 1 | 2.6k |
| xanyun/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | xanyun | 1 | 2.3k |
| q10108158/deepseek-coder-6.7b-instruct-Q5_K_M-GGUF | quantized | GGUF | — | — | other | q10108158 | 0 | 1.2k |
| MrFarhyn/deepseek-coder-6.7b-instruct-Q4_0-GGUF | quantized | Q4_0 | — | — | other | MrFarhyn | 0 | 1k |
| gitarist/deepseek-coder-6.7b-instruct-GGUF | quantized | GGUF | — | — | other | gitarist | 0 | 996 |
| pyarn/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | pyarn | 0 | 964 |
| SkyNotion/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | SkyNotion | 0 | 829 |
| Theta-Lev/deepseek-coder-6.7b-instruct-Q8_0-GGUF | quantized | Q8_0 | — | — | other | Theta-Lev | 0 | 752 |
| anzuo/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | anzuo | 0 | 744 |
| liodon-ai/deepseek-coder-6.7b-instruct-ONNX | quantized | — | — | — | other | liodon-ai | 0 | 679 |
| hellork/deepseek-coder-6.7b-instruct-IQ4_NL-GGUF | quantized | GGUF | — | — | other | hellork | 0 | 570 |
| alphaduriendur/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | alphaduriendur | 0 | 547 |
| Savyasaachin/deepseek-coder-6.7b-instruct-Q5_K_M-GGUF | quantized | GGUF | — | — | other | Savyasaachin | 0 | 522 |
| jamescallander/deepseek-coder-6.7b-instruct_w8a8_g128_rk3588.rkllm | finetune | — | — | — | other | jamescallander | 0 | 481 |
| ori-ai-fabric/ds-trinity-7b-v1 | merge | — | 6.74B | M | — | ori-ai-fabric | 1 | 427 |
| Savyasaachin/deepseek-coder-6.7b-instruct-Q4_K_M-GGUF | quantized | GGUF | — | — | other | Savyasaachin | 0 | 474 |
| louisbrulenaudet/DevPearl-2x7B | merge | — | 11.1B | L | cc-by-sa-4.0 | louisbrulenaudet | 2 | 351 |
| opencsg/opencsg-deepseek-coder-6.7b-pr-review-v0.1 | finetune | — | 6.74B | M | apache-2.0 | opencsg | 2 | 332 |
| ramgpt/deepseek-coder-6.7B-GPTQ | quantized | GPTQ | — | — | apache-2.0 | ramgpt | 0 | 420 |
| SHIBA212121/deepseek-coder-6.7b-instruct-Q5_K_M-GGUF | quantized | GGUF | — | — | other | SHIBA212121 | 0 | 413 |
| didula-wso2/dsc_julia_sftsft_16bit_vllm | finetune | — | 6.74B | M | apache-2.0 | didula-wso2 | 1 | 361 |
| harveymannering/deepseek-coder-6.7b-instruct-finetuned-manimation | finetune | — | — | — | other | harveymannering | 0 | 408 |
| grupo4-bisite/deepseek-coder-6.7b-instruct-Q4_0-GGUF | quantized | Q4_0 | — | — | other | grupo4-bisite | 0 | 404 |
| scthornton/deepseek-coder-6.7b-securecode | adapter | — | — | — | other | scthornton | 1 | 339 |
| Date | Commit | SHA |
|---|---|---|
| 2024-02-02 | Update README.md | e5d64ad |
| 2024-01-05 | Update README.md | bbbe9ae |
| 2024-01-05 | Update README.md | 522ee63 |
| 2023-12-13 | Update tokenizer_config.json | cbb77d7 |
| 2023-12-10 | Update tokenizer_config.json | b56e989 |
| 2023-12-03 | Adding `safetensors` variant of this model (#6) | d7720c2 |
| 2023-11-29 | Update tokenizer_config.json | 88da11d |
| 2023-11-28 | Update tokenizer_config.json | 1fafe1f |
| 2023-11-05 | Update tokenizer_config.json (#2) | 42c3b97 |
| 2023-11-05 | Update generation_config.json | 549dcd6 |
| 2023-11-05 | Update config.json | 263fe02 |
| 2023-11-05 | Upload LICENSE | 73d5f26 |
| 2023-11-05 | Delete LICENSE | 9763180 |
| 2023-11-05 | Update README.md | 7dc2b6c |
| 2023-11-04 | Update tokenizer_config.json | 2aab9dd |
| 2023-11-04 | Update README.md | c32b8df |
| 2023-11-03 | Update README.md | 7ecb2ce |
| 2023-11-02 | Update config.json | f119bee |
| 2023-11-02 | Update README.md | e3d82cd |
| 2023-11-02 | Update README.md | 509e48e |
| 2023-11-01 | Update tokenizer_config.json | 0ed36bf |
| 2023-11-01 | Update README.md | 505e2a1 |
| 2023-11-01 | Update README.md | c050327 |
| 2023-11-01 | change tokenizer | cd9c9d5 |
| 2023-10-29 | upload models | ea05533 |
| 2023-10-29 | upload models | 4a6c861 |
| 2023-10-29 | initial commit | c545484 |
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.
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:38 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:08 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.
