# llama.cpp — User Demand Report

**Week:** 2026-W15
**Generated:** 2026-04-06
**Issues analyzed:** 50 (50 included)
**Need clusters:** 1

## Top 10 User Needs

| Rank | Need | Issues | Score | Category | Examples |
| --- | --- | --- | --- | --- | --- |
| 1 | Vulkan backend stability and Gemma 4 model fixes | 50 | 7.4 | Reliability | [#21483](https://github.com/ggml-org/llama.cpp/issues/21483), [#21473](https://github.com/ggml-org/llama.cpp/issues/21473), [#21446](https://github.com/ggml-org/llama.cpp/issues/21446) |

## Rising Needs

| Need | Rising Score | This Week | Category |
| --- | --- | --- | --- |
| Vulkan backend stability and Gemma 4 model fixes | 51.0x | 50 | Reliability |

## Category Breakdown

- **Reliability**: 1 clusters

## All Need Clusters

### 1. Vulkan backend stability and Gemma 4 model fixes

Users are experiencing multiple crashes and stability issues with the Vulkan backend, particularly when running Gemma 4 models, multimodal vision pipelines, and handling large prompts. These issues include assertion failures, invalid pointer crashes, memory leaks, and incorrect memory estimation that prevent reliable inference. Additionally, users need better documentation for multimodal model configuration and expanded audio modality support.

- **Volume:** 50 issues (39 open, 11 closed)
- **Demand Score:** 7.4
- **Avg Reactions:** 0.4 | **Avg Comments:** 2.6
- **Example issues:** [#21483](https://github.com/ggml-org/llama.cpp/issues/21483), [#21473](https://github.com/ggml-org/llama.cpp/issues/21473), [#21446](https://github.com/ggml-org/llama.cpp/issues/21446), [#21440](https://github.com/ggml-org/llama.cpp/issues/21440), [#21400](https://github.com/ggml-org/llama.cpp/issues/21400)

---

*This report analyzes public GitHub issues only. It represents a signal from public issue discussions, not the full user base.*

*Generated by [ReadYourUsers](https://github.com/study8677/ReadYourUsers)*