Morph MORPH
SIGN IN SIGN UP
dhruvhead / llama.cpp-omni UNCLAIMED
0 0 0 C++

COMMITS

master
July 7, 2026
T
fix(server): drop dead llama-server-impl and llama-server targets (#83)
tc-mb committed
b9d15b8
June 29, 2026
T
fix: swap cert/key args to match SSLServer(cert, key) constructor order (#80)
tc-mb committed
ee12d04
T
fix: handle F32 tensors in ggml_tensor_to_f32 to avoid null to_float segfault (#79)
tc-mb committed
4ad7c49
June 26, 2026
T
feat: support quantized TTS model loading (#75)
tc-mb committed
7aa0108
June 16, 2026
T
feat: support quantized TTS model loading
tc-mb committed
d732c60
T
feat: support quantized TTS model loading (#70)
tc-mb committed
5cf4ef0
June 24, 2026
D
voxcpm2: fix inference correctness, add Metal GPU support and continuation voice cloning (#73)
Dennis Huang committed
dc3163c
June 15, 2026
T
feat: add use_tts flag to session.init for per-session TTS toggle (#69)
tc-mb committed
0f99c52
O
fix: ensure temp_dir is a string in handle_ws_backend function (#68)
oreomaker committed
16354fb
June 13, 2026
T
fix: replace 'common' with 'llama-common' in target_link_libraries (#65)
tc-mb committed
f5771f8
T
Revert "fix: replace 'common' with 'llama-common' in target_link_libraries"
tc-mb committed
eb24d4d
T
fix: replace 'common' with 'llama-common' in target_link_libraries
tc-mb committed
620cce6
June 12, 2026
T
fix: remove duplicate LLM_ARCH_MINICPM4 case in llama_model_rope_type
tc-mb committed
e29787d
T
Feat/omni ws rebase (#63)
tc-mb committed
20dc3b1
T
Feat/voxcpm server (#62)
tc-mb committed
d46153e
June 11, 2026
O
fix: update build targets and references to llama-omni-server in README and source files (#57)
oreomaker committed
3ac8851
T
Feat/vision batch encode (#56)
tc-mb committed
2f1445f
June 5, 2026
T
voxcpm2: restore minicpm4 arch reg + fix ggml_add1/pad/CMake deps dropped by rebase
tc-mb committed
da24197
T
docs: enforce DCO sign-off for every commit
tc-mb committed
4a5318e
T
Opt/r4.1 t2w ane (#52)
tc-mb committed
47f576f
T
gitignore: add apps/, pt/ and output directories
tc-mb committed
402f726
June 4, 2026
T
merge master into feat/voxcpm
tc-mb committed
84b59af
T
merge feat/sync-llama.cpp into master
tc-mb committed
b9ee842
June 3, 2026
T
style: revert formatting noise in llama-model-loader.cpp, keep partial_load changes
tc-mb committed
88d9a85
June 2, 2026
O
fix: improve device management in ggml_metal_device_get and clean up resource handling in audition and token2wav models
oreomaker committed
dd66983
T
fix: avoid sliding during duplex speech and preserve TTS state across chunks
tc-mb committed
1745488
O
feat: add partial_load option to llama_model_loader for flexible tensor loading
oreomaker committed
5623460
June 1, 2026
T
fix: update stream_prefill call to match new signature with text parameter
tc-mb committed
5b870f3
T
server: add standalone omni HTTP server (llama-omni-server)
tc-mb committed
9c21f27
T
fix: add llama_sampler_get_rng API, common_params omni fields, token2wav fixes
tc-mb committed
257dd03