 Georgi Gerganov
|
4301e27319
common : restore grammar-based rejection sampling (#18137)
|
1 mese fa |
 Johannes Gäßler
|
a2c199e479
common: clarify instructions for bug reports (#18134)
|
1 mese fa |
 HonestQiao
|
15dd67d869
model: fix GLM-ASR-Nano-2512 load error (#18130) (#18142)
|
1 mese fa |
 Xuan-Son Nguyen
|
bde461de8c
server: (router) allow child process to report status via stdout (#18110)
|
1 mese fa |
 Piotr Wilkin (ilintar)
|
8faa87db02
Extend run-org-model.py, add (a) batching (b) loading prompt from file (c) multimodal capacity (#18034)
|
1 mese fa |
|
Johannes Gäßler
|
6f1f6a961a
Github: ask for -v logs for params_fit [no ci] (#18128)
|
1 mese fa |
 Alberto Cabrera Pérez
|
669696e00d
ggml-cpu: ARM64: repack version of q8_0 (dotprod and i8mm) (#18096)
|
1 mese fa |
 Tarek Dakhran
|
982060fadc
model: fix LFM2_MOE missing tensors (#18132)
|
1 mese fa |
 Sigbjørn Skjæret
|
6853bee680
ci : clean up webui jobs (#18116)
|
1 mese fa |
 Pascal
|
487674fbb3
common: fix --override-kv to support comma-separated values (#18056)
|
1 mese fa |
 yulo
|
acec774ef6
HIP: Refactor mma for RDNA and CDNA (#17990)
|
1 mese fa |
 Naco Siren
|
5c0d18881e
llama.android : Rewrite Android binding (w/o cpu_features dep) (#17413)
|
1 mese fa |
 TrevorS
|
4b2a4778f8
arg: allow -kvu flag for llama-perplexity (#18117)
|
1 mese fa |
 Aadeshveer Singh
|
58062860af
ggml : use WARP_SIZE/2 for argmax reduction offset (#18092)
|
1 mese fa |
 Yuri Khrustalev
|
2973a65ecb
gguf-py : allow converting multi-tensor models from read-only locations (#18100)
|
1 mese fa |
|
Johannes Gäßler
|
d0794e89d9
llama-fit-params: force disable mlock (#18103)
|
1 mese fa |
|
Johannes Gäßler
|
9dcac6cf9f
llama-fit-params: lower ctx size for multi GPU (#18101)
|
1 mese fa |
|
Johannes Gäßler
|
0e49a7b8b4
llama-fit-params: fix underflow for dense models (#18095)
|
1 mese fa |
|
Johannes Gäßler
|
4164596c76
llama-fit-params: QoL impr. for prints/errors (#18089)
|
1 mese fa |
|
Xuan-Son Nguyen
|
ef83fb8601
model: fix LFM2 missing tensors (#18105)
|
1 mese fa |
|
Johannes Gäßler
|
ec98e20021
llama: fix early stop in params_fit if ctx is set (#18070)
|
1 mese fa |
 yifant-code
|
59977eba7b
server: fix crash when batch > ubatch with embeddings (#17912)
|
1 mese fa |
 Daniel Bevenius
|
79dbae034a
model-conversion : remove -fa option in model card template [no ci] (#18088)
|
1 mese fa |
|
Xuan-Son Nguyen
|
7f2b2f3c77
arch: refactor LLM_TENSOR_NAMES (#18051)
|
1 mese fa |
|
Xuan-Son Nguyen
|
7b1db3d3b7
arg: clarify auto kvu/np being set on server (#17997)
|
1 mese fa |
|
Piotr Wilkin (ilintar)
|
a5251ca11d
Optimization: Qwen3 next autoregressive pass (#17996)
|
1 mese fa |
 Andrew Aladjev
|
fb644247de
CLI: fixed adding cli and completion into docker containers, improved docs (#18003)
|
1 mese fa |
 2114L3
|
5f5f9b4637
server: Update README.md incorrect argument (#18073)
|
1 mese fa |
|
Xuan-Son Nguyen
|
3d86c6c2b5
model: support GLM4V vision encoder (#18042)
|
1 mese fa |
|
Daniel Bevenius
|
9963b81f63
model-conversion : add note about verifying previous models (#18082)
|
1 mese fa |
