Apriel-1.6-15b-Thinker
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 · 32/48 layers on GPU | Fits | ~5.1–6.3 t/s Estimated extrapolated |
| 12 GB GPU + 32 GB RAM | Q4_K_Mllama.cpp CUDAFull GPU | Fits | ~29–31 t/s Estimated |
| 16 GB GPU + 64 GB RAM | Q5_K_Mllama.cpp CUDAFull GPU | Fits | ~52–63 t/s Estimated |
| 24 GB GPU + 64 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~36–45 t/s Estimated |
| 32 GB GPU + 128 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~56–89 t/s Estimated |
| 2× 24 GB GPUs + 128 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~33–42 t/s Estimated |
| 96 GB workstation GPU + 256 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~57–90 t/s Estimated |
| Mac, 24 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~14–15 t/s Estimated |
| Mac, 48 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~14–15 t/s Estimated |
| Mac, 128 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~26–32 t/s Estimated |
| Mac, 512 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~31–45 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 | ~8.6–13 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 bartowski/ServiceNow-AI_Apriel-1.6-15b-Thinker-GGUF | 14.29 GiB | 1 | e77d133d1b | HF files |
| Q5_K_M bartowski/ServiceNow-AI_Apriel-1.6-15b-Thinker-GGUF | 9.56 GiB | 1 | e77d133d1b | HF files |
| Q4_K_M ServiceNow-AI/Apriel-1.6-15b-Thinker-GGUF | 8.18 GiB | 1 | 8cf0961be1 | HF files |
| MLX 8-bit 8-bit cnfusion/Apriel-1.6-15b-Thinker-mlx-8Bit | 14.28 GiB | 3 | 1ae9669a9b | HF files |
| MLX 4-bit 4-bit cnfusion/Apriel-1.6-15b-Thinker-mlx-4Bit | 7.56 GiB | 2 | 8cd9214f84 | HF files |
| Native safetensors BF16 ServiceNow-AI/Apriel-1.6-15b-Thinker | 27.69 GiB | 7 | 5810d5f172 | HF files |
Not found in the representative quant search (not proven unavailable): GGUF FP16, 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
| AA evaluated mode | Index | LiveCodeBench | Terminal-Bench 2.1 | GPQA | HLE | Source |
|---|---|---|---|---|---|---|
| Apriel-v1.6-15B-Thinker AA estimate | 13.4 | — | — | 73.3% | 10.8% | source |
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.