Fresh builds of llama.cpp with AMD ROCm™ 7 acceleration for AMDChat
active 2025-07-25 → 2026-06-24 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 336 days from 2025-07-24 to 2026-06-24. Pushes: 49 total, peak 7 in a day. Pull requests: 28 total, peak 2 in a day. Issues: 35 total, peak 4 in a day. Comments: 91 total, peak 6 in a day. Stars: 134 total, peak 5 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| danielholanda | 113 | 42 | 22 | 37 |
| slojosic-amd | 13 | 3 | 1 | 9 |
| Goldenkoron | 7 | 0 | 0 | 5 |
| ianbmacdonald | 5 | 0 | 0 | 5 |
| Mushoz | 5 | 0 | 0 | 3 |
| Djip007 | 4 | 0 | 0 | 3 |
| jeremyfowers | 3 | 1 | 0 | 1 |
| benrichard-amd | 3 | 0 | 0 | 3 |
| ratcampaign | 3 | 0 | 0 | 1 |
| lhl | 3 | 0 | 0 | 2 |
| iz0eyj | 2 | 0 | 0 | 2 |
| austinsr1 | 2 | 0 | 0 | 0 |
| jchu634 | 2 | 0 | 0 | 1 |
| Geramy | 2 | 0 | 0 | 1 |
| sofiageo | 2 | 2 | 0 | 0 |
| ronmichel | 1 | 0 | 0 | 1 |
| thelittlefireman | 1 | 0 | 1 | 0 |
| boecks | 1 | 0 | 0 | 0 |
| komorebi-c9 | 1 | 0 | 0 | 0 |
| mihairaduonofrei | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#113danielholanda2026-06-16 22:59Broken build since 8 days
- Issue#113danielholanda2026-06-16 22:14Broken build since 8 days
- Pull request#111danielholanda2026-06-12 15:54
- Pull request#103kenvandine2026-06-03 18:20
- Issue#99komorebi-c92026-05-26 14:16[GFX1151] deploying Qwen3.6 using llamacpp error "segmentation fault (core dumped), exit code: -1"
- Issue comment#96realsheox-max2026-05-25 21:28llama.cpp rocm segfaults with R9700 (gfx1201) if integrated GPU (gfx1036) also detected
- Issue comment#85danielholanda2026-05-20 02:11Add missing GPU target for gfx1152, gfx1153
- Issue comment#88danielholanda2026-05-14 23:44Enable LTO to all targets, and apply gfx1151 specific configurations
- Issue comment#53danielholanda2026-05-14 23:42Release Linux artifacts in .tar.gz archives
- Issue comment#84danielholanda2026-05-14 23:42Add .deb packaging for Ubuntu releases
- Issue comment#86danielholanda2026-05-14 23:38gfx1103/780M: Only half of available VRAM is detected
- Issue#86danielholanda2026-05-14 23:38gfx1103/780M: Only half of available VRAM is detected
- Issue#92danielholanda2026-05-14 23:33[gfx1035] device kernel image is invalid
- Issue#89boecks2026-04-26 09:18Add R9700 AI Pro to the list of supported gfx120X cards in the README
- Issue comment#86rockdaboot2026-04-19 21:59gfx1103/780M: Only half of available VRAM is detected
- Pull request#85thelittlefireman2026-04-18 16:20
- Releasedanielholanda2026-04-10 15:19b1236
- Issue comment#75benrichard-amd2026-04-09 04:19qwen3.5-4B gfx 1032 windows | the output no meaning, just a mass
- Releasedanielholanda2026-04-05 15:34b1231
- Releasedanielholanda2026-04-04 14:51b1230
- Issue comment#78danielholanda2026-04-02 22:47llama-rocm Exception Code: 0xC0000005 2026-03-17 22:57:10.991 [Info] (Process) 0x00007FF95FE41F24,
- Issue comment#79danielholanda2026-04-02 05:04ROCm OOM crash on RX 9060 XT (gfx1200) when KV cache exceeds VRAM - Vulkan handles same config gracefully
- Issue comment#80amdjiahangpan2026-04-01 11:26llama-server b1226 hangs at ‘loading model’ Qwen3.5-4B-Q4_K_M on ROCm GPU despite sufficient memory
- Issue comment#19MiklosPathy2026-03-31 02:45Feature Request: Build with rpc-server
- Pull request#77danielholanda2026-03-28 00:48
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 134 stars here means stars gained during the window, not the repo's star count.