Up to date
RadixArk Not declared out 3 months ago
2.25B parameters, text in, text out, 1M tokens of context.
Verified 2026-10-25 Rank #616 in LLMs · score 73.7 method v1 Sources 3 Downloads 30d 3.5M · #42 of 619 tracked
Picked by rule from 29 tracked pieces: the lab's own post, the most shared, a field report, the latest.
Kimi K3 is a 2.8T-parameter open model featuring native vision and a 1-million-token context window.
A fork of Deltafin runs the full 2.8-trillion-parameter Kimi K3 model on a single M5 Max MacBook Pro with 128 GB RAM, streaming 1.45 TB of expert weights from four SSDs.
The 2 picks above, plus these 23. 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: 2 quantized builds and 2 Spaces.
Top Spaces running it: embedl/hfviewer ♥ 181
Weights only, an estimate: leave room for the KV cache. Method
About 1.3 GB at INT4, 2.7 GB at FP8, 5.4 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 (2026-07-27), amber triangles are 10 dated articles (each links to it), grey ticks are repo commits.
2 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.
RadixArk/Kimi-K3-DSpark is a 2.25B parameter DSpark speculator designed for the Kimi K3 target model.[1][2] The model uses a DSparkDraftModel architecture to enable faster inference through speculative decoding.[3][4] It supports context lengths of up to 1 million tokens.[5]
Generated 2026-10-07 15:15 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 | DSparkDraftModel |
| Model type | qwen3 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 2,249,289,601 |
| Size band | M |
| MoE | no |
| Context length | 1,048,576 (config.json max_position_embeddings) |
| License | Not declared |
| Hub snapshot | 2026-10-25 |
vLLM: supported · DSparkDraftModel · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · DSparkDraftModel · 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 Kimi-K3-DSpark: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 5.4 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 2.7 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 1.3 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
quantized: 2
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| Lucebox/Kimi-K3-DSpark-Q8_0-GGUF | quantized | Q8_0 | — | — | — | Lucebox | 1 | 2k |
| Anbeeld/Kimi-K3-DSpark-GGUF | quantized | GGUF | — | — | — | Anbeeld | 0 | 1.1k |
| Date | Commit | SHA |
|---|---|---|
| 2026-08-16 | Link SGLang in the live-target-engine phrase | 3c5bac3 |
| 2026-08-01 | Update README.md | 56ce616 |
| 2026-08-01 | Update README.md | f9d4737 |
| 2026-07-31 | Sync Kimi-K3-DSpark-0731 snapshot and update README | 9c4b257 |
| 2026-07-27 | Add Kimi K3 DSpark model card | eb03982 |
| 2026-07-27 | Duplicate from cm00cm/DSpark-Kimi-K3-V1C | 049288a |
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 21:42 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:15 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.
