Skip to content

Instantly share code, notes, and snippets.

@orimanabu

orimanabu/podman-with-libkrun-for-containerized-AI-workload-acceleration.txt

Created August 6, 2024 14:46
Show Gist options
  • Save orimanabu/e10a4a8544786d8b81095c3607aa0b8d to your computer and use it in GitHub Desktop.
Save orimanabu/e10a4a8544786d8b81095c3607aa0b8d to your computer and use it in GitHub Desktop.
Download ZIP
Podman with libkrun for containerized AI workload acceleration
Raw
podman-with-libkrun-for-containerized-AI-workload-acceleration.txt
% podman run --rm -ti --device /dev/dri -v ~/Downloads:/models:Z quay.io/slopezpa/fedora-vgpu-llama main --temp 0 -m models/Llama-3-ELYZA-JP-8B-q4_k_m.gguf -b 512 -ngl 99 -p "Podmanのlibkrun providerについて教えて下さい"
Log start
main: build = 2238 (56d03d92)
main: built with cc (GCC) 13.2.1 20231205 (Red Hat 13.2.1-6) for aarch64-redhat-linux
main: seed = 1722955285
ggml_vulkan: Found 1 Vulkan devices:
Vulkan0: Virtio-GPU Venus (Apple M2) | uma: 1 | fp16: 1 | warp size: 32
llama_model_loader: loaded meta data with 22 key-value pairs and 291 tensors from models/Llama-3-ELYZA-JP-8B-q4_k_m.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = llama
llama_model_loader: - kv 1: general.name str = Llama-3-8B-optimal-merged-stage2
llama_model_loader: - kv 2: llama.block_count u32 = 32
llama_model_loader: - kv 3: llama.context_length u32 = 8192
llama_model_loader: - kv 4: llama.embedding_length u32 = 4096
llama_model_loader: - kv 5: llama.feed_forward_length u32 = 14336
llama_model_loader: - kv 6: llama.attention.head_count u32 = 32
llama_model_loader: - kv 7: llama.attention.head_count_kv u32 = 8
llama_model_loader: - kv 8: llama.rope.freq_base f32 = 500000.000000
llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 10: general.file_type u32 = 15
llama_model_loader: - kv 11: llama.vocab_size u32 = 128256
llama_model_loader: - kv 12: llama.rope.dimension_count u32 = 128
llama_model_loader: - kv 13: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 14: tokenizer.ggml.pre str = llama-bpe
llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,128256] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 16: tokenizer.ggml.token_type arr[i32,128256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 17: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 128000
llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 128009
llama_model_loader: - kv 20: tokenizer.chat_template str = {% set loop_messages = messages %}{% ...
llama_model_loader: - kv 21: general.quantization_version u32 = 2
llama_model_loader: - type f32: 65 tensors
llama_model_loader: - type q4_K: 193 tensors
llama_model_loader: - type q6_K: 33 tensors
llm_load_vocab: special tokens definition check successful ( 256/128256 ).
llm_load_print_meta: format = GGUF V3 (latest)
llm_load_print_meta: arch = llama
llm_load_print_meta: vocab type = BPE
llm_load_print_meta: n_vocab = 128256
llm_load_print_meta: n_merges = 280147
llm_load_print_meta: n_ctx_train = 8192
llm_load_print_meta: n_embd = 4096
llm_load_print_meta: n_head = 32
llm_load_print_meta: n_head_kv = 8
llm_load_print_meta: n_layer = 32
llm_load_print_meta: n_rot = 128
llm_load_print_meta: n_embd_head_k = 128
llm_load_print_meta: n_embd_head_v = 128
llm_load_print_meta: n_gqa = 4
llm_load_print_meta: n_embd_k_gqa = 1024
llm_load_print_meta: n_embd_v_gqa = 1024
llm_load_print_meta: f_norm_eps = 0.0e+00
llm_load_print_meta: f_norm_rms_eps = 1.0e-05
llm_load_print_meta: f_clamp_kqv = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: n_ff = 14336
llm_load_print_meta: n_expert = 0
llm_load_print_meta: n_expert_used = 0
llm_load_print_meta: rope scaling = linear
llm_load_print_meta: freq_base_train = 500000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_yarn_orig_ctx = 8192
llm_load_print_meta: rope_finetuned = unknown
llm_load_print_meta: model type = 7B
llm_load_print_meta: model ftype = Q4_K - Medium
llm_load_print_meta: model params = 8.03 B
llm_load_print_meta: model size = 4.58 GiB (4.89 BPW)
llm_load_print_meta: general.name = Llama-3-8B-optimal-merged-stage2
llm_load_print_meta: BOS token = 128000 '<|begin_of_text|>'
llm_load_print_meta: EOS token = 128009 '<|eot_id|>'
llm_load_print_meta: LF token = 128 'Ä'
llm_load_tensors: ggml ctx size = 0.22 MiB
llm_load_tensors: offloading 32 repeating layers to GPU
llm_load_tensors: offloading non-repeating layers to GPU
llm_load_tensors: offloaded 33/33 layers to GPU
llm_load_tensors: CPU buffer size = 281.81 MiB
llm_load_tensors: Vulkan0 buffer size = 4403.49 MiB
........................................................................................
llama_new_context_with_model: n_ctx = 512
llama_new_context_with_model: freq_base = 500000.0
llama_new_context_with_model: freq_scale = 1
llama_kv_cache_init: Vulkan0 KV buffer size = 64.00 MiB
llama_new_context_with_model: KV self size = 64.00 MiB, K (f16): 32.00 MiB, V (f16): 32.00 MiB
llama_new_context_with_model: Vulkan_Host input buffer size = 10.01 MiB
llama_new_context_with_model: Vulkan0 compute buffer size = 258.50 MiB
llama_new_context_with_model: Vulkan_Host compute buffer size = 8.00 MiB
llama_new_context_with_model: graph splits (measure): 3
system_info: n_threads = 4 / 4 | AVX = 0 | AVX_VNNI = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 0 | NEON = 1 | ARM_FMA = 1 | F16C = 0 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 0 | SSSE3 = 0 | VSX = 0 | MATMUL_INT8 = 0 |
sampling:
repeat_last_n = 64, repeat_penalty = 1.100, frequency_penalty = 0.000, presence_penalty = 0.000
top_k = 40, tfs_z = 1.000, top_p = 0.950, min_p = 0.050, typical_p = 1.000, temp = 0.000
mirostat = 0, mirostat_lr = 0.100, mirostat_ent = 5.000
sampling order:
CFG -> Penalties -> top_k -> tfs_z -> typical_p -> top_p -> min_p -> temperature
generate: n_ctx = 512, n_batch = 512, n_predict = -1, n_keep = 0
Podmanのlibkrun providerについて教えて下さい。
libkrunはKDEのRun Commandを提供するライブラリです。libkrun providerは、libkrunが提供するRun Commandを実行するためのプロバイダーです。
libkrun providerは、libkrunが提供するRun Commandを実行するために必要な情報を提供します。例えば、実行するコマンドのパスや環境変数などです。
libkrun providerは、KDEのアプリケーションで使用されます。例えば、KDEのファイルマネージャーであるDolphinでは、libkrun providerを使用してRun Commandを実行しています。
以下は、libkrun providerの例です。
```c
#include <krunprovider.h>
class MyKrunProvider : public KRunProvider {
public:
MyKrunProvider(QObject *parent = 0) : KRunProvider(parent) {}
void runCommand(const QUrl &url, const QString &commandLine) override {
// コマンドの実行
QProcess process;
process.start(commandLine);
process.waitForFinished();
}
};
```
この例では、MyKrunProviderクラスがlibkrun providerです。runCommandメソッドは、libkrunから呼び出され、Run Commandを実行するために必要な情報を受け取ります。この例では、QProcessを使用してコマンドを実行しています。 [end of text]
llama_print_timings: load time = 10807.36 ms
llama_print_timings: sample time = 122.81 ms / 307 runs ( 0.40 ms per token, 2499.90 tokens per second)
llama_print_timings: prompt eval time = 4758.71 ms / 12 tokens ( 396.56 ms per token, 2.52 tokens per second)
llama_print_timings: eval time = 27980.78 ms / 306 runs ( 91.44 ms per token, 10.94 tokens per second)
llama_print_timings: total time = 33019.71 ms / 318 tokens
Log end
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment