granite-4.0-micro
Where it runs 8K context · ≥ 5 t/s floor
| Setup | Best plan | Memory | Generate |
|---|---|---|---|
| 8 GB GPU + 32 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | Unknown |
| 12 GB GPU + 32 GB RAM | native BF16vLLMFull GPU | Fits | Unknown |
| 16 GB GPU + 64 GB RAM | native BF16vLLMFull GPU | Fits | Unknown |
| 24 GB GPU + 64 GB RAM | native BF16vLLMFull GPU | Fits | Unknown |
| 32 GB GPU + 128 GB RAM | native BF16vLLMFull GPU | Fits | Unknown |
| 2× 24 GB GPUs + 128 GB RAM | native BF16vLLMFull GPU | Fits | Unknown |
| 96 GB workstation GPU + 256 GB RAM | native BF16vLLMFull GPU | Fits | Unknown |
| Mac, 24 GB | Q8_0llama.cpp MetalUnified memory | Fits | Unknown |
| Mac, 48 GB | Q8_0llama.cpp MetalUnified memory | Fits | Unknown |
| Mac, 128 GB | Q8_0llama.cpp MetalUnified memory | Fits | Unknown |
| Mac, 512 GB | Q8_0llama.cpp MetalUnified memory | Fits | Unknown |
| DGX Spark 128 GB | native BF16vLLMUnified memory | Fits | Unknown |
| Ryzen AI Max+ 395, 128 GB | native BF16vLLMUnified memory | Fits | Unknown |
| CPU only, 64 GB DDR5-5600 dual channel | Q8_0llama.cpp CPUCPU only | Fits | Unknown |
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 ibm-granite/granite-4.0-micro-GGUF | 3.37 GiB | 1 | ec48475f0c | HF files |
| Q5_K_M ibm-granite/granite-4.0-micro-GGUF | 2.27 GiB | 1 | ec48475f0c | HF files |
| Q4_K_M ibm-granite/granite-4.0-micro-GGUF | 1.96 GiB | 1 | ec48475f0c | HF files |
| Native safetensors BF16 ibm-granite/granite-4.0-micro | 6.34 GiB | 2 | 56111ae135 | 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
| AA evaluated mode | Index | LiveCodeBench | Terminal-Bench 2.1 | GPQA | HLE | Source |
|---|---|---|---|---|---|---|
| Granite 4.0 Micro AA estimate | 5.1 | 7.0% | — | 33.6% | 5.0% | 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.