JiRackUltra_14b
Where it runs 8K context · ≥ 5 t/s floor
| Setup | Best plan | Memory | Generate |
|---|---|---|---|
| 8 GB GPU + 32 GB RAM | Q4_K_Mllama.cpp CUDAGPU + CPU layers · 31/48 layers on GPU | Fits | ~4.8–5.9 t/s Estimated extrapolated |
| 12 GB GPU + 32 GB RAM | Q4_K_Mllama.cpp CUDAFull GPU | Fits | ~28–30 t/s Estimated |
| 16 GB GPU + 64 GB RAM | Q5_K_Mllama.cpp CUDAFull GPU | Fits | ~50–62 t/s Estimated |
| 24 GB GPU + 64 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~35–44 t/s Estimated |
| 32 GB GPU + 128 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~55–88 t/s Estimated |
| 2× 24 GB GPUs + 128 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~32–41 t/s Estimated |
| 96 GB workstation GPU + 256 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~56–88 t/s Estimated |
| Mac, 24 GB | Q8_0llama.cpp MetalUnified memory | Fits | ~12–14 t/s Estimated |
| Mac, 48 GB | Q8_0llama.cpp MetalUnified memory | Fits | ~12–14 t/s Estimated |
| Mac, 128 GB | Q8_0llama.cpp MetalUnified memory | Fits | ~19–29 t/s Estimated |
| Mac, 512 GB | Q8_0llama.cpp MetalUnified memory | Fits | ~21–33 t/s Estimated |
| DGX Spark 128 GB | Q8_0llama.cpp CUDAUnified memory | Fits | ~13–14 t/s Estimated |
| Ryzen AI Max+ 395, 128 GB | Q8_0llama.cpp VulkanUnified memory | Fits | ~8.5–12 t/s Estimated |
| CPU only, 64 GB DDR5-5600 dual channel | Q4_K_Mllama.cpp CPUCPU only | Fits | ~2.1–2.5 t/s Estimated extrapolated |
Same plan engine as the planner, with its default reserves and RAM. How plans are chosen.
Files exact Hugging Face bytes
| Artifact | Size | Files | Revision | Source |
|---|---|---|---|---|
| Q8_0 CMSManhattan/JiRackUltra_14b | 14.62 GiB | 1 | 8ea6f82a55 | HF files |
| Q5_K_M CMSManhattan/JiRackUltra_14b | 9.79 GiB | 1 | 8ea6f82a55 | HF files |
| Q4_K_M CMSManhattan/JiRackUltra_14b | 8.37 GiB | 1 | 8ea6f82a55 | HF files |
| Native safetensors BF16 CMSManhattan/JiRackUltra_14b | 27.51 GiB | 1 | 8ea6f82a55 | HF files |
Not found in the representative quant search (not proven unavailable): GGUF FP16, MLX 8-bit, MLX 4-bit, EXL3, AWQ, GPTQ. Complete selected file sets; projectors, MTP/draft and lookup files are listed separately and budgeted only by plans that use them. File bytes are not allocated memory.
Quality
No Artificial Analysis evaluation mapped to this checkpoint.
Arena text rating
Unknown
Publisher-reported benchmarks
None extracted from the model card.
AA and Arena describe the hosted model in the stated mode, not each quantized file; scores are never inherited by base, pruned, abliterated or quantized variants and never averaged together.
How to run it
No model-specific recipe in the research. The planner binds the backend’s documented command template (llama.cpp, Ollama, LM Studio, MLX, vLLM) to your plan: open a plan’s details.
Recipes are source examples (settings are the publisher’s or author’s, not measured optima) and were not executed by LocalFit. A downloadable file does not prove your backend build supports the architecture.
Reported speeds 0
No reported measurement yet. Submit one.