Daniel Hiltgen 33878e671a llama: default qwen2.5vl window attention metadata (#16868) ...

Existing qwen2.5vl GGUFs can contain an empty qwen25vl.vision.fullatt_block_indexes array. The compat layer translated the projector metadata but left clip.vision.n_wa_pattern unset, causing llama-server to fail loading the CLIP model.

Default the runtime compat value to the standard Qwen2.5-VL pattern when the key cannot be derived, and make the converter emit the same default for nil or empty fullatt block metadata.

Fixes #16540

2026-06-24 10:35:29 -07:00

..

sentencepiece

chore(all): replace instances of interface with any (#10067)

2025-04-02 09:44:27 -07:00

testdata

convert: import support for command-r models from safetensors (#6063)

2025-01-15 16:31:22 -08:00

convert_bert.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_commandr.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_deepseek2.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_deepseekocr.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_embeddinggemma_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_embeddinggemma.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gemma2_adapter.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_gemma2.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_gemma3_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gemma3.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gemma3n_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gemma3n.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gemma4_test.go

Add support for gemma4 (#15214)

2026-04-02 11:33:33 -07:00

convert_gemma4.go

Add support for gemma4 (#15214)

2026-04-02 11:33:33 -07:00

convert_gemma.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_glm4moelite_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_glm4moelite.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_glmocr.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gptoss_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_gptoss.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_laguna_test.go

New models (#15861)

2026-04-28 11:50:12 -07:00

convert_laguna.go

New models (#15861)

2026-04-28 11:50:12 -07:00

convert_lfm2_test.go

model: improvements to LFM architectures (#14368)

2026-02-23 14:38:10 -08:00

convert_lfm2_vl_test.go

model: improvements to LFM architectures (#14368)

2026-02-23 14:38:10 -08:00

convert_lfm2_vl.go

model: improvements to LFM architectures (#14368)

2026-02-23 14:38:10 -08:00

convert_lfm2.go

model: improvements to LFM architectures (#14368)

2026-02-23 14:38:10 -08:00

convert_llama4.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_llama_adapter.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_llama_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_llama.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_mistral_causal.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_mistral_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_mistral.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_mixtral.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_mllama.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_nemotron_h_test.go

New models (#15861)

2026-04-28 11:50:12 -07:00

convert_nemotron_h.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_nomicbert.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_olmo.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_phi3.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_qwen2.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_qwen3.go

Add experimental MLX backend and engine with imagegen support (#13648)

2026-01-08 16:18:59 -08:00

convert_qwen3next_test.go

convert: fixes for qwen3next model conversion (#16354)

2026-06-01 09:43:11 -07:00

convert_qwen3next.go

convert: fixes for qwen3next model conversion (#16354)

2026-06-01 09:43:11 -07:00

convert_qwen3vl_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_qwen3vl.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

convert_qwen25vl_test.go

llama: default qwen2.5vl window attention metadata (#16868)

2026-06-24 10:35:29 -07:00

convert_qwen25vl.go

llama: default qwen2.5vl window attention metadata (#16868)

2026-06-24 10:35:29 -07:00

convert_test.go

Add support for gemma4 (#15214)

2026-04-02 11:33:33 -07:00

convert.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

json_compat_test.go

models: add nemotronh architecture support (#14356)

2026-02-22 15:09:14 -08:00

json_compat.go

models: add nemotronh architecture support (#14356)

2026-02-22 15:09:14 -08:00

reader_safetensors.go

New models (#15861)

2026-04-28 11:50:12 -07:00

reader_test.go

New models (#15861)

2026-04-28 11:50:12 -07:00

reader_torch.go

llama4

2025-04-25 16:59:20 -07:00

reader.go

New models (#15861)

2026-04-28 11:50:12 -07:00

sentencepiece_model.proto

all: fix typos in documentation, code, and comments (#7021)

2024-12-10 12:58:06 -08:00

tensor_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

tensor.go

New models (#15861)

2026-04-28 11:50:12 -07:00

tokenizer_spm.go

parsers/renderers: functiongemma (#13521)

2025-12-18 07:55:37 -08:00

tokenizer_test.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00

tokenizer.go

runner: Remove CGO engines, use llama-server exclusively for GGML models (#16031)

2026-05-29 13:35:47 -07:00