Repository Issue Activity (beta)

#803 Canonical vocabulary for fit_level/run_mode/runtime: does the CLI adopt the machine codes?

#638 Detect multiple GPUs

#635 MacBook Neo shows no models under "Good" or "Runnable" due to conservative KV cache filtering

#919 [Bug]: Laptop GPUs inherit their desktop namesake's memory bandwidth (RTX 5070 Laptop -> 672 GB/s on a 128-bit bus), inflating tok/s

#881 test(hardware): add Windows GPU parser fixtures

#914 [Feature]: Recommendations for Tool-Using Coding Agents

#301 Request to add support for NPU and Hybrid models

#617 Question about the origins of this approach

#292 For the given recommendation model, can you provide the corresponding llama.cpp runtime parameters?

#332 Add Gemma4 model family GGUF mappings

#909 [Feature]: Inference bench tui improvements

#876 test(hardware): add mixed multi-GPU fixtures

#901 [Bug]: --json error paths return plain text and inconsistent exit codes

#900 [Bug]: LM Studio installed-detection only sees loaded models (1 of 13)

#887 [Bug]: information is not updated

#791 [Bug]: GGUF models installed by hf cli shown as mlx in TUI

#587 Detect OMLX as an MLX backend and fix MLX filter results

#869 MLX quant suffix stripping lives in two functions with divergent coverage

#874 feat(mcp): add RamaLama runtime discovery

#835 [Bug]: GPU fit grades against total VRAM while CPU paths grade against available RAM, so plan reports a fit on a full card

#883 [Bug]:(short issue description) Formatting in llmfit bench

#880 test(hardware): add Linux sysfs GPU fixtures

#879 test(hardware): add rocm-smi parser fixtures

#878 test(hardware): add system_profiler parser fixtures

#877 test(hardware): add nvidia-smi parser fixtures