Skip to content

server: args for draft model cache types (#11200) - #13782

Merged
ggerganov merged 1 commit into
ggml-org:masterfrom
aa956:server-parameter-draft-cache-type
Jun 19, 2025
Merged

ggerganov merged 1 commit into
ggml-org:masterfrom
aa956:server-parameter-draft-cache-type

Conversation

@aa956

@aa956 aa956 commented May 25, 2025 •

Copy link
Copy Markdown
Contributor

Should fix the #11200, while keeping the default f16 from #10586.

New command line arguments:

Argument Explanation
-ctkd, --cache-type-k-draft TYPE KV cache data type for K for speculative decoding model
allowed values: f32, f16, bf16, q8_0, q4_0, q4_1, iq4_nl, q5_0, q5_1
(default: f16)
(env: LLAMA_ARG_CACHE_TYPE_K_DRAFT)
-ctvd, --cache-type-v-draft TYPE KV cache data type for V for speculative decoding model
allowed values: f32, f16, bf16, q8_0, q4_0, q4_1, iq4_nl, q5_0, q5_1
(default: f16)
(env: LLAMA_ARG_CACHE_TYPE_V_DRAFT)

@CISC

CISC commented Jun 19, 2025

Copy link
Copy Markdown
Member

@ggerganov forgot to merge?

@aa956

aa956 commented Jun 19, 2025

Copy link
Copy Markdown
Contributor Author

@ggerganov forgot to merge?

@ngxson was added automatically on pull request creation, so merging probably needs 2 approvals.

I'm not sure what can I do regarding the failed arm64/ppc64/risc64 builds as I have no access to that kind of hardware and looking at the build logs all of these failed initial apt operations (maybe apt caches/mirrors for these architectures were down at the check time or something similar).

E: Failed to fetch https://security.ubuntu.com/ubuntu/dists/noble-security/main/binary-riscv64/Packages  404  Not Found [IP: 52.252.163.49 80]
E: Some index files failed to download. They have been ignored, or old ones used instead.
Reading package lists...
Building dependency tree...
Reading state information...
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 libcurl4t64:riscv64 : Depends: libgssapi-krb5-2:riscv64 (>= 1.17) but it is not going to be installed
 libssh-4:riscv64 : Depends: libgssapi-krb5-2:riscv64 (>= 1.17) but it is not going to be installed
E: Unable to correct problems, you have held broken packages.

@CISC

CISC commented Jun 19, 2025

Copy link
Copy Markdown
Member

@ggerganov forgot to merge?

@ngxson was added automatically on pull request creation, so merging probably needs 2 approvals.

No, this can be merged now, but just want to make sure it's not on hold for some reason first. :)

I'm not sure what can I do regarding the failed arm64/ppc64/risc64 builds as I have no access to that kind of hardware and looking at the build logs all of these failed initial apt operations (maybe apt caches/mirrors for these architectures were down at the check time or something similar).

Yes, these failures are irrelevant, it has been fixed in CI since.

@ggerganov
ggerganov merged commit d67341d into ggml-org:master Jun 19, 2025
@ggerganov

ggerganov commented Jun 19, 2025 •

Copy link
Copy Markdown
Member

Yes, thanks for the reminder.

qnixsynapse pushed a commit to janhq/llama.cpp that referenced this pull request Jul 2, 2025
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
qnixsynapse pushed a commit to janhq/llama.cpp that referenced this pull request Jul 6, 2025
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
qnixsynapse pushed a commit to janhq/llama.cpp that referenced this pull request Jul 6, 2025
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
qnixsynapse pushed a commit to janhq/llama.cpp that referenced this pull request Jul 10, 2025
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
Seunghhon pushed a commit to Seunghhon/llama.cpp that referenced this pull request Apr 26, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
my-other-github-account pushed a commit to my-other-github-account/llama.cpp that referenced this pull request May 15, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
AlexiAlp pushed a commit to minghaop/llama.cpp that referenced this pull request Jun 2, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
AlexiAlp pushed a commit to minghaop/llama.cpp that referenced this pull request Jun 2, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
fukuro-kun pushed a commit to fukuro-kun/fukuro-llama-cpp-turboquant that referenced this pull request Jul 5, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
zommiommy pushed a commit to zommiommy/llama.cpp that referenced this pull request Aug 18, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
frostyautumnleaf pushed a commit to frostyautumnleaf/llama.cpp that referenced this pull request Oct 5, 2026
…3782)

Co-authored-by: aa956 <27946957+aa956@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants