Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
38 commits
Select commit Hold shift + click to select a range
61cfad0
support for smoldocling
ryan-mangeno Jul 8, 2025
8c18417
fixed merge conflicts
ryan-mangeno Jul 8, 2025
2b209cd
Merge branch 'master' into smoldocling-support
ryan-mangeno Jul 9, 2025
c36703a
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 9, 2025
6ea3b0b
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 9, 2025
7e16cb8
merge conflicts
ryan-mangeno Jul 9, 2025
f4a6fa8
merge conflicts
ryan-mangeno Jul 9, 2025
5050f27
pre tokenizer merge fix
ryan-mangeno Jul 9, 2025
fbfcdae
convert : fix smollm3 jinja template (#14586)
ngxson Jul 9, 2025
43942e6
model : add support for Falcon-H1 family (#14534)
ibrahimkhadraoui Jul 9, 2025
00d2fc2
llama : remove unintended whitespace (#14592)
CISC Jul 9, 2025
e1b0b9e
model : add skt/A.X-4.0 model vocabulary (#14589)
Bing-su Jul 9, 2025
ea265fe
support for smoldocling
ryan-mangeno Jul 8, 2025
529552d
ggml : prevent integer overflow in gguf tensor size calculation (#14595)
Yuuoniy Jul 9, 2025
996c578
fixed merge conflicts
ryan-mangeno Jul 8, 2025
6a3ed2e
rebase and signing
ryan-mangeno Jul 9, 2025
bd137ff
rebase and signing
ryan-mangeno Jul 9, 2025
dd6d36b
Merge branch 'master' into smoldocling-support
ryan-mangeno Jul 9, 2025
c971ea8
Update src/llama-vocab.cpp
ryan-mangeno Jul 9, 2025
ca3cc83
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 9, 2025
b9b532c
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 9, 2025
376ea47
Update src/llama-model.h
ryan-mangeno Jul 9, 2025
5856f8c
safetensors tensor mapping
ryan-mangeno Jul 9, 2025
5c3d948
added back accidental removal of clean spaces for hunyuan
ryan-mangeno Jul 10, 2025
69fa672
Update src/llama-vocab.cpp
ryan-mangeno Jul 10, 2025
f5e4d56
updated hash and reordererd model list
ryan-mangeno Jul 10, 2025
e47dc34
Merge branch 'smoldocling-support' of github.com:ryan-mangeno/llama.c…
ryan-mangeno Jul 10, 2025
037dbbc
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 10, 2025
4331939
Update src/llama-vocab.cpp
ryan-mangeno Jul 10, 2025
661abf5
Update include/llama.h
ryan-mangeno Jul 10, 2025
a3be53c
Update convert_hf_to_gguf.py
ryan-mangeno Jul 10, 2025
859005d
Update convert_hf_to_gguf_update.py
ryan-mangeno Jul 10, 2025
db54ea5
Update src/llama-vocab.cpp
ryan-mangeno Jul 10, 2025
f021d19
removed old tensor name
ryan-mangeno Jul 10, 2025
3f4c532
removed tensor mappings -> handled by smolvlm
ryan-mangeno Jul 10, 2025
310eb46
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 10, 2025
bcf4ad3
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 10, 2025
a4f662d
Update gguf-py/gguf/tensor_mapping.py
ryan-mangeno Jul 10, 2025
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions src/llama-model.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,7 @@ const char * llm_type_name(llm_type type) {
case LLM_TYPE_190M: return "190M";
case LLM_TYPE_220M: return "220M";
case LLM_TYPE_250M: return "250M";
case LLM_TYPE_256M: return "256M";
case LLM_TYPE_270M: return "270M";
case LLM_TYPE_335M: return "335M";
case LLM_TYPE_410M: return "410M";
Expand Down Expand Up @@ -581,6 +582,7 @@ void llama_model::load_hparams(llama_model_loader & ml) {
case 22: type = LLM_TYPE_1B; break;
case 26: type = LLM_TYPE_3B; break;
case 28: type = LLM_TYPE_3B; break; // Llama 3.2 3B
case 30: type = LLM_TYPE_256M; break; // smoldocling 256M
// granite uses a vocab with len 49152
case 32: type = n_vocab == 49152 ? LLM_TYPE_3B : (n_vocab < 40000 ? LLM_TYPE_7B : LLM_TYPE_8B); break;
case 36: type = LLM_TYPE_8B; break; // granite
Expand Down
1 change: 1 addition & 0 deletions src/llama-model.h
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@ enum llm_type {
LLM_TYPE_190M,
LLM_TYPE_220M,
LLM_TYPE_250M,
LLM_TYPE_256M,
LLM_TYPE_270M,
LLM_TYPE_335M,
LLM_TYPE_410M,
Expand Down
2 changes: 2 additions & 0 deletions src/llama-vocab.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -1846,6 +1846,7 @@ void llama_vocab::impl::load(llama_model_loader & ml, const LLM_KV & kv) {
|| t.first == "<EOT>"
|| t.first == "_<EOT>"
|| t.first == "<|end▁of▁sentence|>" // DeepSeek
|| t.first == "<end_of_utterance>" // smoldocling
) {
special_eot_id = t.second;
if ((id_to_token[t.second].attr & LLAMA_TOKEN_ATTR_CONTROL) == 0) {
Expand Down Expand Up @@ -2005,6 +2006,7 @@ void llama_vocab::impl::load(llama_model_loader & ml, const LLM_KV & kv) {
|| t.first == "<EOT>"
|| t.first == "_<EOT>"
|| t.first == "<|end_of_text|>"
|| t.first == "<end_of_utterance>" // smoldocling
) {
special_eog_ids.insert(t.second);
if ((id_to_token[t.second].attr & LLAMA_TOKEN_ATTR_CONTROL) == 0) {
Expand Down
Loading