llama.cpp
llama.cpp copied to clipboard
Published
20 hours ago •
ggml-org
ggml-org
GGML_ASSERT with Vulkan backend and Mixtral 8x7B model
When attempting to load Mixtral 8x7B (Q4_K_M) model with Vulkan backend and any number of layers offloaded to GPU, it fails with GGML_ASSERT. It loads and works with all layers loaded on CPU and this doesn't happen with other models.
Specs
CPU: Ryzen 5 5800H RAM: DDR4 32GB GPU: Radeon RX 6600M 8GB OS: Windows 10 Pro 22H2
Logs
PS C:\Sources\llama.cpp\build\bin\Release> .\main.exe -m "C:\Temp\mixtral-8x7b-instruct-v0.1.Q4_K_M.gguf" -ngl 1
Log start
main: build = 2116 (f026f812)
main: built with MSVC 19.38.33134.0 for x64
main: seed = 1707603695
ggml_vulkan: Found 1 Vulkan devices:
Vulkan0: AMD Radeon RX 6600M | uma: 0 | fp16: 1 | warp size: 64
llama_model_loader: loaded meta data with 26 key-value pairs and 995 tensors from C:\Temp\mixtral-8x7b-instruct-v0.1.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 = mistralai_mixtral-8x7b-instruct-v0.1
llama_model_loader: - kv 2: llama.context_length u32 = 32768
llama_model_loader: - kv 3: llama.embedding_length u32 = 4096
llama_model_loader: - kv 4: llama.block_count u32 = 32
llama_model_loader: - kv 5: llama.feed_forward_length u32 = 14336
llama_model_loader: - kv 6: llama.rope.dimension_count u32 = 128
llama_model_loader: - kv 7: llama.attention.head_count u32 = 32
llama_model_loader: - kv 8: llama.attention.head_count_kv u32 = 8
llama_model_loader: - kv 9: llama.expert_count u32 = 8
llama_model_loader: - kv 10: llama.expert_used_count u32 = 2
llama_model_loader: - kv 11: llama.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 12: llama.rope.freq_base f32 = 1000000.000000
llama_model_loader: - kv 13: general.file_type u32 = 15
llama_model_loader: - kv 14: tokenizer.ggml.model str = llama
llama_model_loader: - kv 15: tokenizer.ggml.tokens arr[str,32000] = ["<unk>", "<s>", "</s>", "<0x00>", "<...
llama_model_loader: - kv 16: tokenizer.ggml.scores arr[f32,32000] = [0.000000, 0.000000, 0.000000, 0.0000...
llama_model_loader: - kv 17: tokenizer.ggml.token_type arr[i32,32000] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ...
llama_model_loader: - kv 18: tokenizer.ggml.bos_token_id u32 = 1
llama_model_loader: - kv 19: tokenizer.ggml.eos_token_id u32 = 2
llama_model_loader: - kv 20: tokenizer.ggml.unknown_token_id u32 = 0
llama_model_loader: - kv 21: tokenizer.ggml.padding_token_id u32 = 0
llama_model_loader: - kv 22: tokenizer.ggml.add_bos_token bool = true
llama_model_loader: - kv 23: tokenizer.ggml.add_eos_token bool = false
llama_model_loader: - kv 24: tokenizer.chat_template str = {{ bos_token }}{% for message in mess...
llama_model_loader: - kv 25: general.quantization_version u32 = 2
llama_model_loader: - type f32: 65 tensors
llama_model_loader: - type f16: 32 tensors
llama_model_loader: - type q8_0: 64 tensors
llama_model_loader: - type q4_K: 833 tensors
llama_model_loader: - type q6_K: 1 tensors
llm_load_vocab: special tokens definition check successful ( 259/32000 ).
llm_load_print_meta: format = GGUF V3 (latest)
llm_load_print_meta: arch = llama
llm_load_print_meta: vocab type = SPM
llm_load_print_meta: n_vocab = 32000
llm_load_print_meta: n_merges = 0
llm_load_print_meta: n_ctx_train = 32768
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 = 8
llm_load_print_meta: n_expert_used = 2
llm_load_print_meta: rope scaling = linear
llm_load_print_meta: freq_base_train = 1000000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_yarn_orig_ctx = 32768
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 = 46.70 B
llm_load_print_meta: model size = 24.62 GiB (4.53 BPW)
llm_load_print_meta: general.name = mistralai_mixtral-8x7b-instruct-v0.1
llm_load_print_meta: BOS token = 1 '<s>'
llm_load_print_meta: EOS token = 2 '</s>'
llm_load_print_meta: UNK token = 0 '<unk>'
llm_load_print_meta: PAD token = 0 '<unk>'
llm_load_print_meta: LF token = 13 '<0x0A>'
llm_load_tensors: ggml ctx size = 0.76 MiB
llm_load_tensors: offloading 1 repeating layers to GPU
llm_load_tensors: offloaded 1/33 layers to GPU
llm_load_tensors: CPU buffer size = 25215.87 MiB
llm_load_tensors: Vulkan0 buffer size = 782.59 MiB
....................................................................................................
llama_new_context_with_model: n_ctx = 512
llama_new_context_with_model: freq_base = 1000000.0
llama_new_context_with_model: freq_scale = 1
llama_kv_cache_init: Vulkan_Host KV buffer size = 62.00 MiB
llama_kv_cache_init: Vulkan0 KV buffer size = 2.00 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 = 9.01 MiB
llama_new_context_with_model: Vulkan0 compute buffer size = 118.83 MiB
llama_new_context_with_model: Vulkan_Host compute buffer size = 118.83 MiB
llama_new_context_with_model: graph splits (measure): 5
GGML_ASSERT: C:\Sources\llama.cpp\ggml-vulkan.cpp:2942: src1 == nullptr || ggml_vk_dim01_contiguous(src1)
I have a very similar experience. Also AMD hardware. Please let us know when Vulkan Support is available in this issue. Thanks a lot
No mixtral over vulkan, and hipBLAS version crashes on dual GPU's :(. Also super waiting for std::cerr << "ggml_vulkan: GGML_OP_MUL_MAT_ID not implemented yet." << std::endl; This to be resolved :)
I'm on it. I hope to get Mixtral working on Vulkan soon. I got all the requirements done, I think.
This issue was closed because it has been inactive for 14 days since being marked as stale.
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}