| | Card: Gyro-M update 2026-10-11 | agentionai | Never scanned |
| | Card: Gyro-S update 2026-10-10 | agentionai |
| | Card: note on the updated MTP draft (upstream names; needs a 2026-10-05+ build) | agentionai |
| | Card v2: headline numbers first, slimmer layout, Strata quick start, LiveCodeBench and MMLU-Pro | agentionai |
| | Card header image | agentionai |
| | Card: CUDA kernel update and crash fix are in agentionai/llama.cpp main (2026-10-04) | agentionai |
| | Card: min-p 0.05; CUDA status and RTX 5090/A6000 numbers (prefill ~800, 128k/160k context); MTP draft fits on 32 GB up to 24k/32k; Ampere CUDA crash workaround | agentionai |
| | Card: two-GPU setup (model on one card, draft on the other); keep --ngram-on-disk | agentionai |
| | Card: footer typo fixes | agentionai |
| | Card: pipeline image-text-to-text; support footer | agentionai |
| | Card: "Which setup?" table per hardware; drafting only on 48 GB+/unified memory; build and platform notes | agentionai |
| | Update README.md | agentionai |
| | Size-class suffixes for the Hub file browser: Gyro-S-TQ1_0, Gyro-M-TQ2_0 | agentionai |
| | Rename the MTP draft to mtp-Qwen3.8-Flash-Next-draft.gguf | agentionai |
| | Card: Gyro-M quality; drop references to the first Gyro file | agentionai |
| | Card: headline separates generating (71) from editing code (86) | agentionai |
| | Card: headline uses the R9700 MTP code numbers | agentionai |
| | Card: R9700 32 GB throughput (tester) and drafting gains | agentionai |
| | Add the vision projector; MTP draft renamed to mtp-* so -hf finds it | agentionai |
| | Add a 2.7 GB MTP draft and faster drafting flags | agentionai |