Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

ci : start using Pythia models over OpenLlama #7470

Merged
merged 6 commits into from
May 23, 2024
Merged

ci : start using Pythia models over OpenLlama #7470

merged 6 commits into from
May 23, 2024

Conversation

ggerganov
Copy link
Owner

depends on #7461

Start using more lightweight models (Pythia 1.4B and 2.8B vs OpenLlama 3B and 7B)

@mofosyne mofosyne added Review Complexity : Medium Generally require more time to grok but manageable by beginner to medium expertise level devops improvements to build systems and github actions labels May 22, 2024
@ggerganov ggerganov force-pushed the gg/ggml-ci-pythia branch from f0d9eda to 57496b2 Compare May 23, 2024 09:54
@ggerganov ggerganov merged commit 55ac3b7 into master May 23, 2024
66 of 77 checks passed
@ggerganov ggerganov deleted the gg/ggml-ci-pythia branch May 23, 2024 12:28
@github-actions GitHub Actions
Copy link
Contributor

📈 llama.cpp server for bench-server-baseline on Standard_NC4as_T4_v3 for phi-2-q4_0: 535 iterations 🚀

Expand details for performance related PR only
  • Concurrent users: 8, duration: 10m
  • HTTP request : avg=8701.94ms p(95)=21251.58ms fails=, finish reason: stop=482 truncated=53
  • Prompt processing (pp): avg=100.69tk/s p(95)=423.2tk/s
  • Token generation (tg): avg=49.42tk/s p(95)=47.08tk/s
  • ggml-org/models/phi-2/ggml-model-q4_0.gguf parallel=8 ctx-size=16384 ngl=33 batch-size=2048 ubatch-size=256 pp=1024 pp+tg=2048 branch=gg/ggml-ci-pythia commit=d2bae455466102487c4c5fce15f37e750c7b5756

prompt_tokens_seconds

More 
---
config:
    xyChart:
        titleFontSize: 12
        width: 900
        height: 600
    themeVariables:
        xyChart:
            titleColor: "#000000"
---
xychart-beta
    title "llama.cpp bench-server-baseline on Standard_NC4as_T4_v3
 duration=10m 535 iterations"
    y-axis "llamacpp:prompt_tokens_seconds"
    x-axis "llamacpp:prompt_tokens_seconds" 1716479076 --> 1716479694
    line [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 702.44, 702.44, 702.44, 702.44, 702.44, 719.54, 719.54, 719.54, 719.54, 719.54, 714.03, 714.03, 714.03, 714.03, 714.03, 728.92, 728.92, 728.92, 728.92, 728.92, 806.66, 806.66, 806.66, 806.66, 806.66, 804.93, 804.93, 804.93, 804.93, 804.93, 791.88, 791.88, 791.88, 791.88, 791.88, 806.38, 806.38, 806.38, 806.38, 806.38, 806.13, 806.13, 806.13, 806.13, 806.13, 802.68, 802.68, 802.68, 802.68, 802.68, 817.45, 817.45, 817.45, 817.45, 817.45, 840.58, 840.58, 840.58, 840.58, 840.58, 785.21, 785.21, 785.21, 785.21, 785.21, 784.11, 784.11, 784.11, 784.11, 784.11, 782.6, 782.6, 782.6, 782.6, 782.6, 769.98, 769.98, 769.98, 769.98, 769.98, 776.75, 776.75, 776.75, 776.75, 776.75, 775.48, 775.48, 775.48, 775.48, 775.48, 774.76, 774.76, 774.76, 774.76, 774.76, 796.49, 796.49, 796.49, 796.49, 796.49, 792.03, 792.03, 792.03, 792.03, 792.03, 795.21, 795.21, 795.21, 795.21, 795.21, 799.09, 799.09, 799.09, 799.09, 799.09, 801.44, 801.44, 801.44, 801.44, 801.44, 799.99, 799.99, 799.99, 799.99, 799.99, 800.7, 800.7, 800.7, 800.7, 800.7, 802.96, 802.96, 802.96, 802.96, 802.96, 819.41, 819.41, 819.41, 819.41, 819.41, 819.96, 819.96, 819.96, 819.96, 819.96, 818.24, 818.24, 818.24, 818.24, 818.24, 818.38, 818.38, 818.38, 818.38, 818.38, 824.71, 824.71, 824.71, 824.71, 824.71, 822.79, 822.79, 822.79, 822.79, 822.79, 824.93, 824.93, 824.93, 824.93, 824.93, 834.88, 834.88, 834.88, 834.88, 834.88, 842.23, 842.23, 842.23, 842.23, 842.23, 848.17, 848.17, 848.17, 848.17, 848.17, 841.45, 841.45, 841.45, 841.45, 841.45, 838.46, 838.46, 838.46, 838.46, 838.46, 838.87, 838.87, 838.87, 838.87, 838.87, 842.3, 842.3, 842.3, 842.3, 842.3, 842.78, 842.78, 842.78, 842.78, 842.78, 833.28, 833.28, 833.28, 833.28, 833.28, 830.98, 830.98, 830.98, 830.98, 830.98, 830.92, 830.92, 830.92, 830.92, 830.92, 830.5, 830.5, 830.5, 830.5, 830.5, 828.3, 828.3, 828.3, 828.3, 828.3, 825.34, 825.34, 825.34, 825.34, 825.34, 830.37, 830.37, 830.37, 830.37, 830.37, 829.68, 829.68, 829.68, 829.68, 829.68, 830.56, 830.56, 830.56, 830.56, 830.56, 834.14, 834.14, 834.14, 834.14, 834.14, 836.78, 836.78, 836.78, 836.78, 836.78, 841.6, 841.6, 841.6, 841.6, 841.6, 841.51, 841.51, 841.51, 841.51, 841.51, 832.53, 832.53, 832.53, 832.53, 832.53, 833.5, 833.5, 833.5, 833.5, 833.5, 833.48, 833.48, 833.48, 833.48, 833.48, 835.2, 835.2, 835.2, 835.2, 835.2, 836.39, 836.39, 836.39, 836.39, 836.39]
Loading
predicted_tokens_seconds
More 
---
config:
    xyChart:
        titleFontSize: 12
        width: 900
        height: 600
    themeVariables:
        xyChart:
            titleColor: "#000000"
---
xychart-beta
    title "llama.cpp bench-server-baseline on Standard_NC4as_T4_v3
 duration=10m 535 iterations"
    y-axis "llamacpp:predicted_tokens_seconds"
    x-axis "llamacpp:predicted_tokens_seconds" 1716479076 --> 1716479694
    line [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 41.96, 41.96, 41.96, 41.96, 41.96, 38.48, 38.48, 38.48, 38.48, 38.48, 34.52, 34.52, 34.52, 34.52, 34.52, 27.89, 27.89, 27.89, 27.89, 27.89, 29.39, 29.39, 29.39, 29.39, 29.39, 29.46, 29.46, 29.46, 29.46, 29.46, 30.46, 30.46, 30.46, 30.46, 30.46, 31.51, 31.51, 31.51, 31.51, 31.51, 31.65, 31.65, 31.65, 31.65, 31.65, 32.0, 32.0, 32.0, 32.0, 32.0, 32.17, 32.17, 32.17, 32.17, 32.17, 32.3, 32.3, 32.3, 32.3, 32.3, 32.51, 32.51, 32.51, 32.51, 32.51, 31.73, 31.73, 31.73, 31.73, 31.73, 31.42, 31.42, 31.42, 31.42, 31.42, 30.55, 30.55, 30.55, 30.55, 30.55, 29.56, 29.56, 29.56, 29.56, 29.56, 29.23, 29.23, 29.23, 29.23, 29.23, 29.5, 29.5, 29.5, 29.5, 29.5, 29.43, 29.43, 29.43, 29.43, 29.43, 29.41, 29.41, 29.41, 29.41, 29.41, 29.77, 29.77, 29.77, 29.77, 29.77, 29.86, 29.86, 29.86, 29.86, 29.86, 30.16, 30.16, 30.16, 30.16, 30.16, 30.09, 30.09, 30.09, 30.09, 30.09, 30.01, 30.01, 30.01, 30.01, 30.01, 30.1, 30.1, 30.1, 30.1, 30.1, 30.16, 30.16, 30.16, 30.16, 30.16, 29.96, 29.96, 29.96, 29.96, 29.96, 30.04, 30.04, 30.04, 30.04, 30.04, 30.45, 30.45, 30.45, 30.45, 30.45, 30.47, 30.47, 30.47, 30.47, 30.47, 30.6, 30.6, 30.6, 30.6, 30.6, 30.71, 30.71, 30.71, 30.71, 30.71, 30.78, 30.78, 30.78, 30.78, 30.78, 30.68, 30.68, 30.68, 30.68, 30.68, 30.52, 30.52, 30.52, 30.52, 30.52, 30.26, 30.26, 30.26, 30.26, 30.26, 29.89, 29.89, 29.89, 29.89, 29.89, 30.15, 30.15, 30.15, 30.15, 30.15, 30.19, 30.19, 30.19, 30.19, 30.19, 30.32, 30.32, 30.32, 30.32, 30.32, 30.41, 30.41, 30.41, 30.41, 30.41, 30.34, 30.34, 30.34, 30.34, 30.34, 29.9, 29.9, 29.9, 29.9, 29.9, 29.82, 29.82, 29.82, 29.82, 29.82, 29.06, 29.06, 29.06, 29.06, 29.06, 28.51, 28.51, 28.51, 28.51, 28.51, 28.5, 28.5, 28.5, 28.5, 28.5, 28.44, 28.44, 28.44, 28.44, 28.44, 28.55, 28.55, 28.55, 28.55, 28.55, 28.56, 28.56, 28.56, 28.56, 28.56, 28.66, 28.66, 28.66, 28.66, 28.66, 28.66, 28.66, 28.66, 28.66, 28.66, 28.68, 28.68, 28.68, 28.68, 28.68, 28.73, 28.73, 28.73, 28.73, 28.73, 28.69, 28.69, 28.69, 28.69, 28.69, 28.87, 28.87, 28.87, 28.87, 28.87, 28.9, 28.9, 28.9, 28.9, 28.9, 29.07, 29.07, 29.07, 29.07, 29.07]
Loading

Details

kv_cache_usage_ratio

More 
---
config:
    xyChart:
        titleFontSize: 12
        width: 900
        height: 600
    themeVariables:
        xyChart:
            titleColor: "#000000"
---
xychart-beta
    title "llama.cpp bench-server-baseline on Standard_NC4as_T4_v3
 duration=10m 535 iterations"
    y-axis "llamacpp:kv_cache_usage_ratio"
    x-axis "llamacpp:kv_cache_usage_ratio" 1716479076 --> 1716479694
    line [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.19, 0.19, 0.19, 0.19, 0.19, 0.49, 0.49, 0.49, 0.49, 0.49, 0.32, 0.32, 0.32, 0.32, 0.32, 0.12, 0.12, 0.12, 0.12, 0.12, 0.18, 0.18, 0.18, 0.18, 0.18, 0.24, 0.24, 0.24, 0.24, 0.24, 0.15, 0.15, 0.15, 0.15, 0.15, 0.14, 0.14, 0.14, 0.14, 0.14, 0.18, 0.18, 0.18, 0.18, 0.18, 0.18, 0.18, 0.18, 0.18, 0.18, 0.17, 0.17, 0.17, 0.17, 0.17, 0.21, 0.21, 0.21, 0.21, 0.21, 0.32, 0.32, 0.32, 0.32, 0.32, 0.25, 0.25, 0.25, 0.25, 0.25, 0.46, 0.46, 0.46, 0.46, 0.46, 0.37, 0.37, 0.37, 0.37, 0.37, 0.27, 0.27, 0.27, 0.27, 0.27, 0.15, 0.15, 0.15, 0.15, 0.15, 0.16, 0.16, 0.16, 0.16, 0.16, 0.11, 0.11, 0.11, 0.11, 0.11, 0.14, 0.14, 0.14, 0.14, 0.14, 0.19, 0.19, 0.19, 0.19, 0.19, 0.18, 0.18, 0.18, 0.18, 0.18, 0.14, 0.14, 0.14, 0.14, 0.14, 0.18, 0.18, 0.18, 0.18, 0.18, 0.21, 0.21, 0.21, 0.21, 0.21, 0.13, 0.13, 0.13, 0.13, 0.13, 0.32, 0.32, 0.32, 0.32, 0.32, 0.16, 0.16, 0.16, 0.16, 0.16, 0.15, 0.15, 0.15, 0.15, 0.15, 0.12, 0.12, 0.12, 0.12, 0.12, 0.19, 0.19, 0.19, 0.19, 0.19, 0.19, 0.19, 0.19, 0.19, 0.19, 0.17, 0.17, 0.17, 0.17, 0.17, 0.26, 0.26, 0.26, 0.26, 0.26, 0.26, 0.26, 0.26, 0.26, 0.26, 0.42, 0.42, 0.42, 0.42, 0.42, 0.33, 0.33, 0.33, 0.33, 0.33, 0.11, 0.11, 0.11, 0.11, 0.11, 0.09, 0.09, 0.09, 0.09, 0.09, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.32, 0.32, 0.32, 0.32, 0.32, 0.58, 0.58, 0.58, 0.58, 0.58, 0.58, 0.58, 0.58, 0.58, 0.58, 0.61, 0.61, 0.61, 0.61, 0.61, 0.43, 0.43, 0.43, 0.43, 0.43, 0.17, 0.17, 0.17, 0.17, 0.17, 0.29, 0.29, 0.29, 0.29, 0.29, 0.27, 0.27, 0.27, 0.27, 0.27, 0.14, 0.14, 0.14, 0.14, 0.14, 0.2, 0.2, 0.2, 0.2, 0.2, 0.28, 0.28, 0.28, 0.28, 0.28, 0.23, 0.23, 0.23, 0.23, 0.23, 0.18, 0.18, 0.18, 0.18, 0.18, 0.24, 0.24, 0.24, 0.24, 0.24, 0.13, 0.13, 0.13, 0.13, 0.13, 0.07, 0.07, 0.07, 0.07, 0.07, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13, 0.13]
Loading
requests_processing
More 
---
config:
    xyChart:
        titleFontSize: 12
        width: 900
        height: 600
    themeVariables:
        xyChart:
            titleColor: "#000000"
---
xychart-beta
    title "llama.cpp bench-server-baseline on Standard_NC4as_T4_v3
 duration=10m 535 iterations"
    y-axis "llamacpp:requests_processing"
    x-axis "llamacpp:requests_processing" 1716479076 --> 1716479694
    line [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 5.0, 5.0, 5.0, 5.0, 5.0, 7.0, 7.0, 7.0, 7.0, 7.0, 8.0, 8.0, 8.0, 8.0, 8.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 6.0, 6.0, 6.0, 6.0, 6.0, 4.0, 4.0, 4.0, 4.0, 4.0, 3.0, 3.0, 3.0, 3.0, 3.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 5.0, 5.0, 5.0, 5.0, 5.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 5.0, 5.0, 5.0, 5.0, 5.0, 7.0, 7.0, 7.0, 7.0, 7.0, 8.0, 8.0, 8.0, 8.0, 8.0, 5.0, 5.0, 5.0, 5.0, 5.0, 7.0, 7.0, 7.0, 7.0, 7.0, 7.0, 7.0, 7.0, 7.0, 7.0, 4.0, 4.0, 4.0, 4.0, 4.0, 8.0, 8.0, 8.0, 8.0, 8.0, 3.0, 3.0, 3.0, 3.0, 3.0, 4.0, 4.0, 4.0, 4.0, 4.0, 2.0, 2.0, 2.0, 2.0, 2.0, 5.0, 5.0, 5.0, 5.0, 5.0, 3.0, 3.0, 3.0, 3.0, 3.0, 5.0, 5.0, 5.0, 5.0, 5.0, 7.0, 7.0, 7.0, 7.0, 7.0, 4.0, 4.0, 4.0, 4.0, 4.0, 5.0, 5.0, 5.0, 5.0, 5.0, 0.0, 0.0, 0.0, 0.0, 0.0, 6.0, 6.0, 6.0, 6.0, 6.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 5.0, 5.0, 5.0, 5.0, 5.0, 6.0, 6.0, 6.0, 6.0, 6.0, 8.0, 8.0, 8.0, 8.0, 8.0, 6.0, 6.0, 6.0, 6.0, 6.0, 4.0, 4.0, 4.0, 4.0, 4.0, 0.0, 0.0, 0.0, 0.0, 0.0, 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 5.0, 6.0, 6.0, 6.0, 6.0, 6.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 8.0, 6.0, 6.0, 6.0, 6.0, 6.0, 5.0, 5.0, 5.0, 5.0, 5.0, 4.0, 4.0, 4.0, 4.0, 4.0, 2.0, 2.0, 2.0, 2.0, 2.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 7.0, 7.0, 7.0, 7.0, 7.0, 6.0, 6.0, 6.0, 6.0, 6.0, 5.0, 5.0, 5.0, 5.0, 5.0, 7.0, 7.0, 7.0, 7.0, 7.0, 6.0, 6.0, 6.0, 6.0, 6.0]
Loading

teleprint-me pushed a commit to teleprint-me/llama.cpp that referenced this pull request May 23, 2024
@ggerganov @teleprint-me
ci : use Pythia models instead of OpenLlama (ggerganov#7470)
3a3f28b 
* ci : start using Pythia models over OpenLlama

ggml-ci

* ci : disable q2_k ppl tests

* ci : use convert-hf-to-gguf.py

* ci : update gg_get_model

* ci : fix convert outfile name

ggml-ci

* llama : gptneox arch use F32 attn prec

ggml-ci
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
devops improvements to build systems and github actions Review Complexity : Medium Generally require more time to grok but manageable by beginner to medium expertise level
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@ggerganov @mofosyne