Qabbel
DeepSeek R1 Distill Qwen 1.5B vs DeepSeek-R1-Distill-Qwen-32B
Qiegħed żewġ mudelli sa erbgħa ma' xulxin fuq l-oqsma li jiddeċiedu deployment: kuntest, prezz, reġjun, liċenzja, lingwi, hosting, u x'hu verifikat.
| DeepSeek R1 Distill Qwen 1.5B | DeepSeek-R1-Distill-Qwen-32B | |
|---|---|---|
| Fornitur | DeepSeek | DeepSeek |
| Parametri | 1.8B | 32.76B |
| Tieqa tal-kuntest | Mhux verifikat: il-fornitur ma ddokumentax din id-dikjarazzjoni. | 131k |
| Liċenzja | mit | mit |
| Modalitajiet | text | text |
| Data tar-rilaxx | 20 Jan 2025 | 20 Jan 2025 |
| Lingwi | Mhux verifikat: il-fornitur ma ddokumentax din id-dikjarazzjoni. | Qawwi: en, zh Adekwat: de, fr, es, it, pt, ru |
| Piżijiet miftuħa | iva | iva |
| Kompiti | Raġunament, Spiża baxxa | Raġunament, Ikkodifikar, Spiża baxxa, Sommarizzazzjoni |
Fejn żewġ mudelli jvarjaw, din it-tabella turi d-differenza u tieqaf. LLM EU ma tippubblikax badge ta' rebbieħ mingħajr punteġġ dokumentat għal dik il-kompitu, u figura nieqsa tibqa' nieqsa minflok ma ssir żero.
Meta tużah
DeepSeek R1 Distill Qwen 1.5B
Saħħiet
- open weights are published on Hugging Face and can be downloaded without a vendor agreement
- mit licence, so commercial self-hosting is permitted without a separate agreement
- at 1.8B parameters it fits on a single accelerator
Limiti
- catalogue only — LLM EU does not run this model, and nothing on this card implies an LLM EU endpoint
- no evaluation has been run by LLM EU
- the published configuration records no context window, so the card carries none
- at 1.8B parameters, the weights fit on a single modest accelerator
Kompiti
DeepSeek-R1-Distill-Qwen-32B
Saħħiet
- fits a single 80GB accelerator
- MIT licence
- retains the R1 reasoning style that smaller base models usually lack
Limiti
- a distillation, so quality is bounded by the teacher
- long reasoning traces increase output cost
- no evaluation has been run by LLM EU
Kompiti
Kompiti: Raġunament, Spiża baxxa