SAVRN Model Hub · Comparisons
Kimi-K3-DSpark vs Llama-3.2-3B-Instruct
Kimi-K3-DSpark has 2.2B parameters and Llama-3.2-3B-Instruct has 3.2B parameters; at 16-bit, Kimi-K3-DSpark needs about 5.4 GB (1x MI300X from $1.85 an hour) and Llama-3.2-3B-Instruct about 7.7 GB (1x MI300X from $1.85 an hour).
| Field | Kimi-K3-DSpark RadixArk/Kimi-K3-DSpark | Llama-3.2-3B-Instruct meta-llama/Llama-3.2-3B-Instruct |
|---|---|---|
| Publisher | RadixArk | Meta Llama |
| Task | Text generation | Text generation |
| Modality | Text | Text |
| Parameters, as reported | 2.2B parameters | 3.2B parameters |
| Architecture | DSparkDraftModel | LlamaForCausalLM |
| Library | transformers | transformers |
| Context length | 1,048,576 tokens | Not stated |
| Repository size | 4.5 GB | 12.9 GB |
| Artifact formats | safetensors | safetensors, pytorch |
| License | Not stated | llama3.2 |
| Access | Open weights, no gate | Access requested at publisher |
| Memory at 16-bit (weights and margin) | 5.4 GB | 7.7 GB |
| Cheapest GPUs at 16-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Memory at 4-bit (weights and margin) | 1.3 GB | 1.9 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 3c5bac301d9c | 0cb88a4f764b |
| Downloads reported by the hub | 6.3M | 1.6M |
| Last observed | 2026-10-09 | 2026-10-09 |
An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.
SAVRN's Notes on Kimi-K3-DSpark
Read this as a companion, not a model you serve on its own. RadixArk trained Kimi-K3-DSpark as a DSpark speculator for the Kimi K3 target, a 2.2 billion parameter draft built for faster inference through speculative decoding. Five layers and 4.5 GB of weights keep it light: 5.4 GB of memory at 16-bit, 2.7 GB at 8-bit, 1.3 GB at 4-bit, and one MI300X with 192 GB at $1.85 an hour on-demand is the cheapest listing we track. That covers the draft only; the Kimi K3 target sizes your box.
The license field is empty, so get written terms from RadixArk before deployment; open access is not permission to run it commercially. Match your stack too: the checkpoint was trained with SpecForge on hidden states from a live SGLang target engine, and the 1,048,576 token context is the draft's ceiling, so the target needs the same window.
SAVRN's Notes on Llama-3.2-3B-Instruct
3.2B parameters in 6.4 GB of safetensors — the published 12.9 GB also carries an original/consolidated.00.pth copy, so you can shed half the download. Memory runs 7.7 GB at 16-bit, 3.9 at 8-bit, 1.9 at 4-bit; one MI300X at $1.85 an hour covers every precision, and at these sizes the card serves thousands of concurrent small chats. The economics are dominated by everything but the weights.
The license is llama3.2 and access is requested at the publisher — accept the terms before download; it is not a standard permissive grant. The page states no context length and no evaluations on our record. The natural neighbors are the fine-tunes listed nearby and Qwen's instruct models of similar size; pick Llama when your toolchain or compliance stack is already Meta-shaped, otherwise an Apache-2.0 alternative removes a legal review step.
Questions
Which is larger, Kimi-K3-DSpark or Llama-3.2-3B-Instruct?
Llama-3.2-3B-Instruct (3.2B parameters) is larger than Kimi-K3-DSpark (2.2B parameters), by the parameter counts their publishers report.
Which is cheaper to run, Kimi-K3-DSpark or Llama-3.2-3B-Instruct?
At 4-bit, Kimi-K3-DSpark fits on 1x MI300X from $1.85 an hour and Llama-3.2-3B-Instruct on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.