K2-Think-V2 Specialist
Where it runs 8K context · ≥ 5 t/s floor
| Setup | Best plan | Memory | Generate |
|---|---|---|---|
| 8 GB GPU + 32 GB RAM | Does not fit | ||
| 12 GB GPU + 32 GB RAM | Does not fit | ||
| 16 GB GPU + 64 GB RAM | Does not fit | ||
| 24 GB GPU + 64 GB RAM | Does not fit | ||
| 32 GB GPU + 128 GB RAM | Does not fit | ||
| 2× 24 GB GPUs + 128 GB RAM | Does not fit | ||
| 96 GB workstation GPU + 256 GB RAM | native F32vLLMGPU streams weights from RAM · 179 GiB in RAM | Fits | ≤ 0.2 t/s Ceiling |
| Mac, 24 GB | Does not fit | ||
| Mac, 48 GB | Does not fit | ||
| Mac, 128 GB | MLX 4-bitMLX-LMUnified memory | Fits | ~11–13 t/s Estimated |
| Mac, 512 GB | MLX 4-bitMLX-LMUnified memory | Fits | ~13–18 t/s Estimated |
| DGX Spark 128 GB | Does not fit | ||
| Ryzen AI Max+ 395, 128 GB | Does not fit | ||
| CPU only, 64 GB DDR5-5600 dual channel | Does not fit | ||
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 |
|---|---|---|---|---|
| MLX 4-bit 4-bit McG-221/K2-Think-V2-mxfp4-gs32 | 35.90 GiB | 8 | 70fa6b274c | HF files |
| Native safetensors F32 IFM/K2-Think-V2 | 270 GiB | 62 | f86e5d11c5 | HF files |
Not found in the representative quant search (not proven unavailable): GGUF FP16, GGUF Q8_0, GGUF Q5_K_M, GGUF Q4_K_M, MLX 8-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 |
|---|---|---|---|---|---|---|
| K2 Think V2 AA estimate | 11.5 | 57.0% | 15.0% | 71.3% | 10.1% | 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.