Jesus Gulfo b0dcb8192b server: fix speculation after an image (#28715) ...

* server: fix speculation after an image

Pass the actual position to the drafter after an image, instead of the
token count. Affects every drafter, not just DFlash.

* rename draft n_past to pos0

n_past is used to denote number of tokens and this parameter is meant to be a position

2026-09-11 11:33:26 +03:00

..

batched

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

batched.swift

examples : remove references to make in examples [no ci] (#15457)

2025-08-21 06:12:28 +02:00

convert-llama2c-to-ggml

fix: check gguf array type before reading (#27075)

2026-08-15 11:45:30 +02:00

debug

common: fix missing exports in llama-common (#22340)

2026-04-27 08:06:39 +03:00

deprecation-warning

Fix locale-dependent float printing in GGUF metadata (#17331)

2026-03-04 09:30:40 +01:00

diffusion

args: refactor mlock/mmap/directio into load-mode (#20834)

2026-07-23 20:32:56 +08:00

embedding

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

eval-callback

common: fix missing exports in llama-common (#22340)

2026-04-27 08:06:39 +03:00

gen-docs

server: add get_info tool (#26522)

2026-08-03 18:51:02 +02:00

gguf

Fix locale-dependent float printing in GGUF metadata (#17331)

2026-03-04 09:30:40 +01:00

gguf-hash

build : fix xcframework + cmake clean-up (#27304)

2026-08-18 11:16:51 +03:00

idle

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

llama-eval

llama-eval : fix crash when answer is None in HTML dump (#25435)

2026-07-08 10:00:03 +03:00

llama.android

cmake : add install() for impl libraries + fix apple builds (#23511)

2026-05-22 11:46:26 +03:00

llama.swiftui

llama : deprecate llama_kv_self_ API (#14030)

2025-06-06 14:11:15 +03:00

lookahead

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

lookup

llama : support multi-output backend sampling (#25532)

2026-08-10 16:58:56 +03:00

model-conversion

requirements: use stable torch packages on s390x (#26864)

2026-08-11 21:58:53 +08:00

parallel

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

passkey

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

retrieval

libs : rename libcommon -> libllama-common (#21936)

2026-04-17 11:11:46 +03:00

simple

Fix locale-dependent float printing in GGUF metadata (#17331)

2026-03-04 09:30:40 +01:00

simple-chat

Fix locale-dependent float printing in GGUF metadata (#17331)

2026-03-04 09:30:40 +01:00

simple-cmake-pkg

examples : add missing code block end marker [no ci] (#17756)

2025-12-04 14:17:30 +01:00

speculative

llama : support multi-output backend sampling (#25532)

2026-08-10 16:58:56 +03:00

speculative-simple

server: fix speculation after an image (#28715)

2026-09-11 11:33:26 +03:00

sycl

[SYCL] support OP OPT_STEP_ADAMW, OPT_STEP_SGD (#25268)

2026-08-17 07:31:29 +03:00

test-cmake

examples : print ggml_version and ggml_commit in test-cmake [no ci] (#28538)

2026-09-07 12:11:40 +02:00

training

finetune: fix no KV cache (#27199)

2026-09-02 23:53:32 +02:00

CMakeLists.txt

tests : move save-load-state from examples to tests (#23336)

2026-05-21 14:41:50 +03:00

convert_legacy_llama.py

convert : minor fixes for numpy 2.x (#23571)

2026-05-24 09:51:31 +02:00

json_schema_pydantic_example.py

py : type-check all Python scripts with Pyright (#8341)

2024-07-07 15:04:39 -04:00

json_schema_to_grammar.py

common/json-schema : fix GBNF grammar generation for empty object schemas (#28279)

2026-09-03 15:37:50 -05:00

llama.vim

chore : correct typos [no ci] (#20041)

2026-03-05 08:50:21 +01:00

pydantic_models_to_grammar_examples.py

llama : move end-user examples to tools directory (#13249)

2025-05-02 20:27:13 +02:00

pydantic_models_to_grammar.py

ci : bump ty to 0.0.78 (#28548)

2026-09-07 21:10:06 +02:00

reason-act.sh

scripts : make the shell scripts cross-platform (#14341)

2025-06-30 10:17:18 +02:00

regex_to_grammar.py

py : switch to snake_case (#8305)

2024-07-05 07:53:33 +03:00

server_embd.py

llama : fix FA when KV cache is not used (i.e. embeddings) (#12825)

2025-04-08 19:54:51 +03:00

server-llama2-13B.sh

scripts : make the shell scripts cross-platform (#14341)

2025-06-30 10:17:18 +02:00

ts-type-to-grammar.sh

scripts : make the shell scripts cross-platform (#14341)

2025-06-30 10:17:18 +02:00