@FIR-757: Update SDK to 0.1.4 and update release to 0.0.3 for tsi-ggml #20
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The test results with ./run_llama_cli.sh with 5 tokens is as follows
+++
root@agilex7_dk_si_agf014ea:/usr/bin/tsi/v0.1.1.tsv31_06_06_2025/bin# ./run_llama_cli.sh
my cat's name is Max. He'
llama_perf_sampler_print: sampling time = 111.70 ms / 11 runs ( 10.15 ms per token, 98.47 tokens per second)llama_perf_context_print: load time = 132926.48 ms
llama_perf_context_print: prompt eval time = 109957.33 ms / 6 tokens (18326.22 ms per token, 0.05 tokens per second)
llama_perf_context_print: eval time = 195682.91 ms / 4 runs (48920.73 ms per token, 0.02 tokens per second)
llama_perf_context_print: total time = 328764.01 ms / 10 tokens
GGML Tsavorite Profiling Results:
------------------------------------------------------------------------------------------------------------------------
Calls Total(ms) T/call Self(ms) Function
------------------------------------------------------------------------------------------------------------------------
33160 100086.000 3.018 47907.157 [32%] RuntimeHostShim::awaitCommandListCompletion
18920 29912.952 1.581 29912.952 └─ [10%] [ txe_silu ]
14080 22010.102 1.563 22010.102 └─ [ 7%] [ txe_mult ]
160 253.071 1.582 253.071 └─ [ 0%] [ txe_add ]
33160 1.178 0.000 1.178 └─ [ 0%] TXE 0 Idle
1 114.000 114.000 18.000 [ 0%] GGML Tsavorite
1 96.000 96.000 96.000 └─ [ 0%] RuntimeHostShim::initialize
1 52.000 52.000 52.000 [ 0%] RuntimeHostShim::finalize
33160 26.000 0.001 26.000 [ 0%] RuntimeHostShim::loadBlob
33160 23.000 0.001 23.000 [ 0%] RuntimeHostShim::finalizeCommandList
33160 5.000 0.000 5.000 [ 0%] RuntimeHostShim::addCommandToList
33161 3.000 0.000 3.000 [ 0%] RuntimeHostShim::allocate
33160 3.000 0.000 3.000 [ 0%] RuntimeHostShim::createCommandList
113720 0.000 0.000 0.000 [ 0%] RuntimeHostShim::getShmemManager
33160 0.000 0.000 0.000 [ 0%] RuntimeHostShim::launchBlob
33160 0.000 0.000 0.000 [ 0%] RuntimeHostShim::unloadBlob
33160 0.000 0.000 0.000 [ 0%] RuntimeHostShim::deallocate
========================================================================================================================
412163 308849.000 0.749308849.000 [100%] TOTAL
========================================================================================================================
root@agilex7_dk_si_agf014ea:/usr/bin/tsi/v0.1.1.tsv31_06_06_2025/bin#
+++
atrivedi-tsavoritesi
requested review from
DashingR,
LewisLui777,
akapoor3518,
dineshReddy6381,
dmpatra,
gkethamallax,
mmankal and
reach2shaunak
mmankal
approved these changes
Jun 18, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The test results with ./run_llama_cli.sh with 5 tokens is as follows
+++
root@agilex7_dk_si_agf014ea:/usr/bin/tsi/v0.1.1.tsv31_06_06_2025/bin# ./run_llama_cli.sh
my cat's name is Max. He'
llama_perf_sampler_print: sampling time = 111.70 ms / 11 runs ( 10.15 ms per token, 98.47 tokens per second)llama_perf_context_print: load time = 132926.48 ms
llama_perf_context_print: prompt eval time = 109957.33 ms / 6 tokens (18326.22 ms per token, 0.05 tokens per second)
llama_perf_context_print: eval time = 195682.91 ms / 4 runs (48920.73 ms per token, 0.02 tokens per second)
llama_perf_context_print: total time = 328764.01 ms / 10 tokens
GGML Tsavorite Profiling Results:
Calls Total(ms) T/call Self(ms) Function
33160 100086.000 3.018 47907.157 [32%] RuntimeHostShim::awaitCommandListCompletion
18920 29912.952 1.581 29912.952 └─ [10%] [ txe_silu ]
14080 22010.102 1.563 22010.102 └─ [ 7%] [ txe_mult ]
160 253.071 1.582 253.071 └─ [ 0%] [ txe_add ]
33160 1.178 0.000 1.178 └─ [ 0%] TXE 0 Idle
1 114.000 114.000 18.000 [ 0%] GGML Tsavorite
1 96.000 96.000 96.000 └─ [ 0%] RuntimeHostShim::initialize
1 52.000 52.000 52.000 [ 0%] RuntimeHostShim::finalize
33160 26.000 0.001 26.000 [ 0%] RuntimeHostShim::loadBlob
33160 23.000 0.001 23.000 [ 0%] RuntimeHostShim::finalizeCommandList
33160 5.000 0.000 5.000 [ 0%] RuntimeHostShim::addCommandToList
33161 3.000 0.000 3.000 [ 0%] RuntimeHostShim::allocate
33160 3.000 0.000 3.000 [ 0%] RuntimeHostShim::createCommandList
113720 0.000 0.000 0.000 [ 0%] RuntimeHostShim::getShmemManager
33160 0.000 0.000 0.000 [ 0%] RuntimeHostShim::launchBlob
33160 0.000 0.000 0.000 [ 0%] RuntimeHostShim::unloadBlob
33160 0.000 0.000 0.000 [ 0%] RuntimeHostShim::deallocate
412163 308849.000 0.749308849.000 [100%] TOTAL
root@agilex7_dk_si_agf014ea:/usr/bin/tsi/v0.1.1.tsv31_06_06_2025/bin# +++
Make sure to read the contributing guidelines before submitting a PR