Olmo-3-7B-Think Specialist
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 · 31/32 layers on GPU | Fits | ~20–28 t/s Estimated extrapolated |
| 12 GB GPU + 32 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~28–30 t/s Estimated |
| 16 GB GPU + 64 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~58–70 t/s Estimated |
| 24 GB GPU + 64 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~58–73 t/s Estimated |
| 32 GB GPU + 128 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~91–146 t/s Estimated |
| 2× 24 GB GPUs + 128 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~53–69 t/s Estimated |
| 96 GB workstation GPU + 256 GB RAM | Q8_0llama.cpp CUDAFull GPU | Fits | ~94–147 t/s Estimated |
| Mac, 24 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~23–24 t/s Estimated |
| Mac, 48 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~23–24 t/s Estimated |
| Mac, 128 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~43–51 t/s Estimated |
| Mac, 512 GB | MLX 8-bitMLX-LMUnified memory | Fits | ~51–74 t/s Estimated |
| DGX Spark 128 GB | Q8_0llama.cpp CUDAUnified memory | Fits | ~21–23 t/s Estimated |
| Ryzen AI Max+ 395, 128 GB | Q8_0llama.cpp VulkanUnified memory | Fits | ~14–20 t/s Estimated |
| CPU only, 64 GB DDR5-5600 dual channel | Q4_K_Mllama.cpp CPUCPU only | Fits | ~3.1–3.6 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/allenai_Olmo-3-7B-Think-GGUF | 7.23 GiB | 1 | d5a1bf052b | HF files |
| Q5_K_M bartowski/allenai_Olmo-3-7B-Think-GGUF | 4.85 GiB | 1 | d5a1bf052b | HF files |
| Q4_K_M bartowski/allenai_Olmo-3-7B-Think-GGUF | 4.16 GiB | 1 | d5a1bf052b | HF files |
| MLX 8-bit 8-bit mlx-community/Olmo-3-7B-Think-8bit | 7.22 GiB | 2 | 3365ab060d | HF files |
| MLX 4-bit 4-bit mlx-community/Olmo-3-7B-Think-4bit | 3.82 GiB | 1 | 65294083d6 | HF files |
| Native safetensors BF16 allenai/Olmo-3-7B-Think | 13.59 GiB | 3 | d97e442d7c | 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 |
|---|---|---|---|---|---|---|
| Olmo 3 7B Think AA estimate | 5.6 | 0.0% | — | 51.6% | 6.0% | 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.