* Fix music generation token stopping for quantized models
In Phase 1 lyrics mode, the FSM transitions to CODES state after
TOKEN_THINK_END and disables itself. The quantized Q4_K_M model was
not efficiently generating TOKEN_IM_END to stop the generation,
causing it to continue until hitting the 8192 token limit.
This fix forces TOKEN_IM_END to be generated immediately after
TOKEN_THINK_END in lyrics mode, ensuring clean completion of the
planning phase without excessive token generation.
Testing shows generation now completes in ~500ms instead of 80+
seconds with timeout errors.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Clarify comment - fix applies to all models, not just quantized
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Improve fix: only force TOKEN_IM_END at token limit
Instead of forcing TOKEN_IM_END immediately after TOKEN_THINK_END,
only force it when we've reached the token limit. This allows the model
to generate lyrics after the thinking block while still preventing KV
cache exhaustion.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
* Apply suggestions from code review
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* metal:add conv_3d backend
Rebased with master and resolved conflicts.
* Resolved issues related to changes in variable names
* kernel void kernel_upscale_bilinear_f32 was missing in my branch, added back, should pass all tests now
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
ACL graph capture disallows host-to-device memcpy and device memory
malloc/free on the captured stream. Pre-load the RoPE cache before
capture so that:
- Host-to-device copies and allocations run on the non-captured stream
- Cache metadata is populated and memory pool is warmed up
- During capture, only on-device computations are recorded; host-side
and allocation branches are skipped
* fix(openvino): explicit memset in buffer_context allocation
* minor
---------
Co-authored-by: Dan Hoffman <dhoffman@cyket.net>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* added support for internvl's dynamic high-resolution (Qianfan-OCR needed)
* add min/max dynamic patch to gguf meta
* clean up
* simplified handling min/max dynamic patch
* reuse llava_uhd logic for slice images
* provide default values for older models
* flake8
* prevent writing 0 value to gguf
* remove duplicated resolution candidates with a better algorithm
* fix indentation
* format
* add protection from divide by zero
* change to 0 to be safe
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
* ggml-cuda: native bf16 flash attention for vec and tile kernels
mma kernel still converts bf16 to fp16 before launch, native mma bf16 todo
* ggml-cuda: address code owner review feedback
reverted tile kernel changes to avoid larger refactor
* fix ci failures on turing and hip
* fix bf16 vec kernel compile on hip v_dot2 platforms
* add comments
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
* Increase per-thread work if the K-dimension is small
With tensor parallelism, the K-dimension of the FFN-down matrices is split, which makes it quite small, especially for MOEs. For example, Qwen3-30b-A3B has a K-dimension of 768, and Qwen3235B-A22B has k-dimension of 1536.
The current heuristic uses a group of 4 warps irrespective of K-dimension size, resulting in some of the threads being idle. This results in poor performance for these matrices.
This change increases the number of output elements per block for such cases.
* Limit this change to ncols_dst = 1
* tab to space
* misc : prefer ggml-org models in docs and examples
Prefer referring to known-good quantizations under ggml-org rather than
3rd-party uploaders.
* remove accidentally committed file
* grammar: add test case for nullable symbol loop
Reproduce stack overflow (or OOM) with ( [x]* )* found while adding
GBNF support to ripgrep-edit.
llama-server reproducer:
curl \
-X POST \
-d '{
"messages": [{ "role": "user", "content": "write yes" }],
"grammar": "root ::= ( [x]* )*"
}' \
-H "Content-Type: application/json" \
http://localhost:8811/v1/chat/completions
* grammar: prevent stack overflow with nullable symbol loop
Fix a potential stack overflow in llama_grammar_advance_stack that
could occur when processing grammars with nullable symbols that lead
to infinite derivations of empty strings. The fix introduces cycle
detection by tracking visited stacks to prevent infinite recursion.
rg-edit regexp: llama_grammar_advance_stack
rg-edit extra-args: -A20
rg-edit directive: """Rewrite: fix the following segfault:
[..]
⚫ Testing segfault. Grammar:
root ::= ( [x]* )*
root ::= ( [x]* )*
Segmentation fault build/bin/test-grammar-integration"""
gptel-context:
(("~/llama.cpp/src/llama-grammar.cpp")
("~/llama.cpp/tests/test-grammar-integration.cpp")
("~/llama.cpp/grammars/./list.gbnf")
("~/llama.cpp/grammars/./json_arr.gbnf")
("~/llama.cpp/grammars/./json.gbnf")
("~/llama.cpp/grammars/./japanese.gbnf")
("~/llama.cpp/grammars/./english.gbnf")
("~/llama.cpp/grammars/./chess.gbnf")
("~/llama.cpp/grammars/./c.gbnf")
("~/llama.cpp/grammars/./arithmetic.gbnf")
("~/llama.cpp/grammars/./README.md"))
* grammar: convert recursive llama_grammar_advance_stack to iterative
This change converts the function to an iterative approach using
explicit stacks, which prevents deep recursion and eliminates the risk
of stack overflow.
rg-edit regexp: llama_grammar_advance_stack
rg-edit extra-args: -A30
rg-edit directive: """Rewrite: fix the following segfault:
[..]
⚫ Testing segfault. Grammar:
root ::= ( [x]* )*
root ::= ( [x]* )*
Segmentation fault build/bin/test-grammar-integration
convert from recursive to interactive"""
gptel-context:
(("~/llama.cpp/src/llama-grammar.cpp")
("~/llama.cpp/tests/test-grammar-integration.cpp")
("~/llama.cpp/grammars/./list.gbnf")
("~/llama.cpp/grammars/./json_arr.gbnf")
("~/llama.cpp/grammars/./json.gbnf")
("~/llama.cpp/grammars/./japanese.gbnf")
("~/llama.cpp/grammars/./english.gbnf")
("~/llama.cpp/grammars/./chess.gbnf")
("~/llama.cpp/grammars/./c.gbnf")
("~/llama.cpp/grammars/./arithmetic.gbnf")
("~/llama.cpp/grammars/./README.md"))
v2: Added a `std::set` to perform tree-based lookups with O(N log N)
complexity. Testing with a parallel run of `test-grammar-integration`
shows a double-digit percentage increase in runtime. An
`unordered_set` with O(1) hashing was also evaluated, but the overhead
of constructing hash keys from pointers made it significantly slower
than the rbtree implementation that only requires an ordering
operator. The performance regression in the test suite appears
justified by the overall reduction in algorithmic complexity.
Co-developed-by: Piotr Wilkin (ilintar) <piotr.wilkin@syndatis.com>
* grammar: add test case for hang in repetition grammar processing
This commit adds a new test case to the grammar integration tests that
specifically targets a hang scenario in the repetition grammar parser
found while adding GBNF support to ripgrep-edit.
llama-server reproducer:
curl \
-X POST \
-d '{
"messages": [{ "role": "user", "content": "write yes" }],
"grammar": "root ::= (([^x]*){0,99}){0,99}"
}' \
-H "Content-Type: application/json" \
http://localhost:8811/v1/chat/completions
* grammar: add repetition threshold check
The change introduces a maximum repetition threshold to avoid
excessive rule expansion during grammar parsing. When parsing
repetition patterns like {m,n}, the parser now calculates the
potential number of rules that would be generated and throws an error
if the product of previous rules and new rules exceeds the threshold.
A test case was added to verify the threshold is properly enforced for
deeply nested repetition patterns that would otherwise cause hangs.
The MEAN/CLS/LAST pooling paths in encode() and decode() used
n_embd_inp() (16384 for qwen3vl with deepstack) to read from the
pooled embedding tensor, which only has n_embd_out() (4096) floats
per sequence. This caused a tensor read out of bounds assertion.
Fixes embedding mode for Qwen3-VL-Embedding models.
rpc : prevent division by zero in deserialize_tensor
When receiving an RPC message with a deprecated tensor type (e.g., type 4 or 5 where `blck_size == 0`), `ggml_row_size()` will trigger a division by zero (SIGFPE) and crash the rpc-server.
This patch adds a simple validation check in `deserialize_tensor` to return `nullptr` if the requested tensor type has a block size of 0.
(Note: This was originally reported via Security Advisory and maintainer suggested dropping a patch here).
* style: remove trailing whitespace
Davi Henrique Linhares
Concedo
Aleksander Grygier
Geo Maciolek
Georgi Gerganov
Eric Zhang
Xuan-Son Nguyen
Pascal
Wagner Bruna
Alistair Stewart
Rashid Ul Islam
Jhen-Jie Hong
Chenguang Li
Dan Hoffman
shaofeiqi
bssrdf
DorianRudolph
Johannes Gäßler
Sigbjørn Skjæret
Evgeny Kurnevsky
Neo Zhang
Patrick Buckley
Gaurav Garg
ddh0
Andrea Arcangeli
Tom Hillbrunner
y198
Michael Wand
Matt Corallo