Skip to content

Issues: ggml-org/llama.cpp

examples : add configuration presets
#10932 opened Dec 21, 2024 by ggerganov
Open 3
changelog : libllama API
#9289 opened Sep 3, 2024 by ggerganov
Open 9
changelog : llama-server REST API
#9291 opened Sep 3, 2024 by ggerganov
Open 14
Beta
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

333 Open 4,691 Closed
333 Open 4,691 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Compile bug: paths with spaces fail on Unix with Vulkan backend bug-unconfirmed
#13288 opened May 3, 2025 by kangalio
1
Eval bug: b5237 broke Llama Scout bug Something isn't working
#13287 opened May 3, 2025 by steampunque
@JohannesGaessler
3
Misc. bug: Completions hang after CUDA error, but health endpoint reports all OK bug-unconfirmed
#13281 opened May 3, 2025 by lee-b
Eval bug: grammar breaks sampling in Qwen3 MoE bug-unconfirmed
#13280 opened May 3, 2025 by matteoserva
8
Misc. bug: llama-server webui overriding command line parameters bug-unconfirmed
#13277 opened May 3, 2025 by merc4derp
Feature Request: Granite 4 Support enhancement New feature or request
#13275 opened May 2, 2025 by gabe-l-hart
5 of 16 tasks
2
Feature Request: add to llama-bench device info reporting of "bf16:1", if built with VK_KHR_bfloat16 support and driver also supports it.. enhancement New feature or request
#13274 opened May 2, 2025 by oscarbg
4 tasks done
Feature Request: add per-request "reasoning" options in llama-server enhancement New feature or request
#13272 opened May 2, 2025 by ngxson
Compile bug: nvcc fatal : Unsupported gpu architecture 'compute_120' bug-unconfirmed
#13271 opened May 2, 2025 by jacekpoplawski
6
Misc. bug: Server does not always cancel requests for disconnected connections bug-unconfirmed
#13262 opened May 2, 2025 by CyberShadow
Misc. bug: the output file of llama-quantize is not gguf format bug-unconfirmed
#13258 opened May 2, 2025 by samsosu
Eval bug: sentencepiece tokenizer generates incorrect tokens bug-unconfirmed
#13256 opened May 2, 2025 by taylorchu
1
Eval bug: ggml_cuda_compute_forward: MUL_MAT failed when using FA + MLA on DeepSeekv3 0324, on mixed CPU + GPU bug-unconfirmed
#13252 opened May 2, 2025 by Panchovix
1
When using the qwen2.5-vl model on AMD Ryzen APU under Windows, the error "failed to allocate Vulkan0 buffer of size 4342230552" may appear. bug-unconfirmed
#13250 opened May 2, 2025 by xeden3
Misc. bug: terminate called after throwing an instance of 'vk::DeviceLostError' bug-unconfirmed
#13248 opened May 1, 2025 by Som-anon
Feature Request: Support multimodal LLMs such as Qwen2.5-VL as embedding models enhancement New feature or request
#13247 opened May 1, 2025 by cebtenzzre
4 tasks done
3
Feature Request: s390x CI enhancement New feature or request
#13243 opened May 1, 2025 by taronaeo
4 tasks done
1
Feature Request: Allow disabling offload_op for backends by user enhancement New feature or request
#13241 opened May 1, 2025 by hjc4869
4 tasks done
1
Eval bug: -sm row causes GGML_ASSERT fail in Llama 4 Scout bug-unconfirmed
#13240 opened May 1, 2025 by FullstackSensei
1
Feature Request: XiaomiMiMo/MiMo-7B-RL enhancement New feature or request
#13218 opened Apr 30, 2025 by Superluckyhu
4 tasks done
1
1
Eval bug: Qwen3-30B-A3B-Q4_K_M: Vulkan ~10% slower than AVX2 bug-unconfirmed
#13217 opened Apr 30, 2025 by ZUIcat
3
Eval bug: Qwen3 30B A3B is slow with CUDA bug-unconfirmed
#13211 opened Apr 30, 2025 by Nepherpitou
1
Eval bug: Can't utilize all 16 threads / 8 CPU cores for prompt processing when using llama-server. works fine with llama-cli bug-unconfirmed
#13197 opened Apr 29, 2025 by c777s
2
Eval bug: Persistent <think> Tags in Qwen3-32B Output Despite enable_thinking: False and --reasoning-format none in llama.cpp bug-unconfirmed
#13189 opened Apr 29, 2025 by shyn01
1
Misc. bug: rpc-server crash without cache bug-unconfirmed
#13185 opened Apr 29, 2025 by hbuxiaofei
1
ProTip! Find all open issues with in progress development work with linked:pr.