llama.cpp build: 716301d1 (5757)
Meta-Llama-3.1-70B-Instruct-IQ4_XS.gguf
./build/bin/llama-batched-bench -ngl 999 -m ../Meta-Llama-3.1-70B-Instruct-IQ4_XS.gguf -fa -npl 1,2,4,6,8,10,12 -npp 512 -ntg 128 -c 16384
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: Orin, compute capability 8.7, VMM: yes
build: 5757 (716301d1) with cc (Ubuntu 13.3.0-6ubuntu2~24.04) 13.3.0 for aarch64-linux-gnu
llama_model_load_from_file_impl: using device CUDA0 (Orin) - 58814 MiB free
llama_model_loader: loaded meta data with 33 key-value pairs and 724 tensors from ../Meta-Llama-3.1-70B-Instruct-IQ4_XS.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.type str = model
llama_model_loader: - kv 2: general.name str = Meta Llama 3.1 70B Instruct
llama_model_loader: - kv 3: general.finetune str = Instruct
llama_model_loader: - kv 4: general.basename str = Meta-Llama-3.1
llama_model_loader: - kv 5: general.size_label str = 70B
llama_model_loader: - kv 6: general.license str = llama3.1
llama_model_loader: - kv 7: general.tags arr[str,6] = ["facebook", "meta", "pytorch", "llam...
llama_model_loader: - kv 8: general.languages arr[str,8] = ["en", "de", "fr", "it", "pt", "hi", ...
llama_model_loader: - kv 9: llama.block_count u32 = 80
llama_model_loader: - kv 10: llama.context_length u32 = 131072
llama_model_loader: - kv 11: llama.embedding_length u32 = 8192
llama_model_loader: - kv 12: llama.feed_forward_length u32 = 28672
llama_model_loader: - kv 13: llama.attention.head_count u32 = 64
llama_model_loader: - kv 14: llama.attention.head_count_kv u32 = 8
llama_model_loader: - kv 15: llama.rope.freq_base f32 = 500000.000000
llama_model_loader: - kv 16: llama.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 17: general.file_type u32 = 30
llama_model_loader: - kv 18: llama.vocab_size u32 = 128256
llama_model_loader: - kv 19: llama.rope.dimension_count u32 = 128
llama_model_loader: - kv 20: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 21: tokenizer.ggml.pre str = llama-bpe
llama_model_loader: - kv 22: tokenizer.ggml.tokens arr[str,128256] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 23: tokenizer.ggml.token_type arr[i32,128256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 24: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv 25: tokenizer.ggml.bos_token_id u32 = 128000
llama_model_loader: - kv 26: tokenizer.ggml.eos_token_id u32 = 128009
llama_model_loader: - kv 27: tokenizer.chat_template str = {{- bos_token }}\n{%- if custom_tools ...
llama_model_loader: - kv 28: general.quantization_version u32 = 2
llama_model_loader: - kv 29: quantize.imatrix.file str = /models_out/Meta-Llama-3.1-70B-Instru...
llama_model_loader: - kv 30: quantize.imatrix.dataset str = /training_dir/calibration_datav3.txt
llama_model_loader: - kv 31: quantize.imatrix.entries_count i32 = 560
llama_model_loader: - kv 32: quantize.imatrix.chunks_count i32 = 125
llama_model_loader: - type f32: 162 tensors
llama_model_loader: - type q5_K: 80 tensors
llama_model_loader: - type q6_K: 1 tensors
llama_model_loader: - type iq4_xs: 481 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = IQ4_XS - 4.25 bpw
print_info: file size = 35.29 GiB (4.30 BPW)
load: special tokens cache size = 256
load: token to piece cache size = 0.7999 MB
print_info: arch = llama
print_info: vocab_only = 0
print_info: n_ctx_train = 131072
print_info: n_embd = 8192
print_info: n_layer = 80
print_info: n_head = 64
print_info: n_head_kv = 8
print_info: n_rot = 128
print_info: n_swa = 0
print_info: is_swa_any = 0
print_info: n_embd_head_k = 128
print_info: n_embd_head_v = 128
print_info: n_gqa = 8
print_info: n_embd_k_gqa = 1024
print_info: n_embd_v_gqa = 1024
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-05
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 28672
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: causal attn = 1
print_info: pooling type = 0
print_info: rope type = 0
print_info: rope scaling = linear
print_info: freq_base_train = 500000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 131072
print_info: rope_finetuned = unknown
print_info: ssm_d_conv = 0
print_info: ssm_d_inner = 0
print_info: ssm_d_state = 0
print_info: ssm_dt_rank = 0
print_info: ssm_dt_b_c_rms = 0
print_info: model type = 70B
print_info: model params = 70.55 B
print_info: general.name = Meta Llama 3.1 70B Instruct
print_info: vocab type = BPE
print_info: n_vocab = 128256
print_info: n_merges = 280147
print_info: BOS token = 128000 '<|begin_of_text|>'
print_info: EOS token = 128009 '<|eot_id|>'
print_info: EOT token = 128009 '<|eot_id|>'
print_info: EOM token = 128008 '<|eom_id|>'
print_info: LF token = 198 'Ċ'
print_info: EOG token = 128001 '<|end_of_text|>'
print_info: EOG token = 128008 '<|eom_id|>'
print_info: EOG token = 128009 '<|eot_id|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors: offloading 80 repeating layers to GPU
load_tensors: offloading output layer to GPU
load_tensors: offloaded 81/81 layers to GPU
load_tensors: CUDA0 model buffer size = 35606.98 MiB
load_tensors: CPU_Mapped model buffer size = 532.31 MiB
..................................................................................................
llama_context: constructing llama_context
llama_context: n_seq_max = 12
llama_context: n_ctx = 16384
llama_context: n_ctx_per_seq = 1365
llama_context: n_batch = 2048
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = 1
llama_context: freq_base = 500000.0
llama_context: freq_scale = 1
llama_context: n_ctx_per_seq (1365) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CUDA_Host output buffer size = 5.87 MiB
llama_kv_cache_unified: CUDA0 KV buffer size = 5120.00 MiB
llama_kv_cache_unified: size = 5120.00 MiB ( 16384 cells, 80 layers, 12 seqs), K (f16): 2560.00 MiB, V (f16): 2560.00 MiB
llama_context: CUDA0 compute buffer size = 266.50 MiB
llama_context: CUDA_Host compute buffer size = 52.13 MiB
llama_context: graph nodes = 2567
llama_context: graph splits = 2
main: n_kv_max = 16384, n_batch = 2048, n_ubatch = 512, flash_attn = 1, is_pp_shared = 0, n_gpu_layers = 999, n_threads = 12, n_threads_batch = 12
| PP | TG | B | N_KV | T_PP s | S_PP t/s | T_TG s | S_TG t/s | T s | S t/s |
|-------|--------|------|--------|----------|----------|----------|----------|----------|----------|
| 512 | 128 | 1 | 640 | 14.184 | 36.10 | 42.285 | 3.03 | 56.468 | 11.33 |
| 512 | 128 | 2 | 1280 | 8.941 | 114.53 | 50.120 | 5.11 | 59.060 | 21.67 |
| 512 | 128 | 4 | 2560 | 17.983 | 113.88 | 71.604 | 7.15 | 89.587 | 28.58 |
| 512 | 128 | 6 | 3840 | 27.168 | 113.07 | 81.013 | 9.48 | 108.181 | 35.50 |
| 512 | 128 | 8 | 5120 | 36.480 | 112.28 | 93.166 | 10.99 | 129.646 | 39.49 |
| 512 | 128 | 10 | 6400 | 45.901 | 111.54 | 95.976 | 13.34 | 141.877 | 45.11 |
| 512 | 128 | 12 | 7680 | 55.496 | 110.71 | 78.255 | 19.63 | 133.751 | 57.42 |
llama_perf_context_print: load time = 148042.28 ms
llama_perf_context_print: prompt eval time = 718462.74 ms / 27408 tokens ( 26.21 ms per token, 38.15 tokens per second)
llama_perf_context_print: eval time = 42280.79 ms / 128 runs ( 330.32 ms per token, 3.03 tokens per second)
llama_perf_context_print: total time = 866615.25 ms / 27536 tokens ./build/bin/llama-bench -m ../Meta-Llama-3.1-70B-Instruct-IQ4_XS.gguf -ngl 999 -fa 1 -n 0 -p 64,128,256,512
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: Orin, compute capability 8.7, VMM: yes
| model | size | params | backend | ngl | fa | test | t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| llama 70B IQ4_XS - 4.25 bpw | 35.29 GiB | 70.55 B | CUDA | 999 | 1 | pp64 | 71.54 ± 0.06 |
| llama 70B IQ4_XS - 4.25 bpw | 35.29 GiB | 70.55 B | CUDA | 999 | 1 | pp128 | 109.25 ± 0.26 |
| llama 70B IQ4_XS - 4.25 bpw | 35.29 GiB | 70.55 B | CUDA | 999 | 1 | pp256 | 91.70 ± 46.86 |
| llama 70B IQ4_XS - 4.25 bpw | 35.29 GiB | 70.55 B | CUDA | 999 | 1 | pp512 | 113.74 ± 0.19 |
build: 716301d1 (5757)