llama_model_load: error loading model: vk::PhysicalDevice::createDevice: ErrorDeviceLost

[BreakShoot]

What happened?

How to reproduce

1. Install latest vulkan release (llama-b3889-bin-win-vulkan-x64)
2. Install model
3. Place model in /models
4. Run given example command
   llama-cli -m models/jina.gguf --prompt "Once upon a time"

Output

build: 3889 (b6d6c528) with MSVC 19.29.30154.0 for x64
main: llama backend init
main: load the model and apply lora adapter, if any
llama_model_loader: loaded meta data with 31 key-value pairs and 196 tensors from models/jina.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              = jina-bert-v2
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                               general.name str              = Jina Bert Implementation
llama_model_loader: - kv   3:                       general.organization str              = Jinaai
llama_model_loader: - kv   4:                         general.size_label str              = 137M
llama_model_loader: - kv   5:                            general.license str              = apache-2.0
llama_model_loader: - kv   6:                               general.tags arr[str,4]       = ["sentence-transformers", "feature-ex...
llama_model_loader: - kv   7:                          general.languages arr[str,1]       = ["en"]
llama_model_loader: - kv   8:                           general.datasets arr[str,1]       = ["allenai/c4"]
llama_model_loader: - kv   9:                   jina-bert-v2.block_count u32              = 12
llama_model_loader: - kv  10:                jina-bert-v2.context_length u32              = 8192
llama_model_loader: - kv  11:              jina-bert-v2.embedding_length u32              = 768
llama_model_loader: - kv  12:           jina-bert-v2.feed_forward_length u32              = 3072
llama_model_loader: - kv  13:          jina-bert-v2.attention.head_count u32              = 12
llama_model_loader: - kv  14:  jina-bert-v2.attention.layer_norm_epsilon f32              = 0.000000
llama_model_loader: - kv  15:                          general.file_type u32              = 17
llama_model_loader: - kv  16:              jina-bert-v2.attention.causal bool             = false
llama_model_loader: - kv  17:                  jina-bert-v2.pooling_type u32              = 1
llama_model_loader: - kv  18:            tokenizer.ggml.token_type_count u32              = 2
llama_model_loader: - kv  19:                       tokenizer.ggml.model str              = bert
llama_model_loader: - kv  20:                         tokenizer.ggml.pre str              = jina-v2-en
llama_model_loader: - kv  21:                      tokenizer.ggml.tokens arr[str,30528]   = ["[PAD]", "[unused0]", "[unused1]", "...
llama_model_loader: - kv  22:                  tokenizer.ggml.token_type arr[i32,30528]   = [3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  23:            tokenizer.ggml.unknown_token_id u32              = 100
llama_model_loader: - kv  24:          tokenizer.ggml.seperator_token_id u32              = 102
llama_model_loader: - kv  25:            tokenizer.ggml.padding_token_id u32              = 0
llama_model_loader: - kv  26:                tokenizer.ggml.cls_token_id u32              = 101
llama_model_loader: - kv  27:               tokenizer.ggml.mask_token_id u32              = 103
llama_model_loader: - kv  28:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  29:               tokenizer.ggml.add_eos_token bool             = true
llama_model_loader: - kv  30:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:  111 tensors
llama_model_loader: - type q5_K:   72 tensors
llama_model_loader: - type q6_K:   13 tensors
llm_load_vocab: special tokens cache size = 5
llm_load_vocab: token to piece cache size = 0.2032 MB
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = jina-bert-v2
llm_load_print_meta: vocab type       = WPM
llm_load_print_meta: n_vocab          = 30528
llm_load_print_meta: n_merges         = 0
llm_load_print_meta: vocab_only       = 0
llm_load_print_meta: n_ctx_train      = 8192
llm_load_print_meta: n_embd           = 768
llm_load_print_meta: n_layer          = 12
llm_load_print_meta: n_head           = 12
llm_load_print_meta: n_head_kv        = 12
llm_load_print_meta: n_rot            = 64
llm_load_print_meta: n_swa            = 0
llm_load_print_meta: n_embd_head_k    = 64
llm_load_print_meta: n_embd_head_v    = 64
llm_load_print_meta: n_gqa            = 1
llm_load_print_meta: n_embd_k_gqa     = 768
llm_load_print_meta: n_embd_v_gqa     = 768
llm_load_print_meta: f_norm_eps       = 1.0e-12
llm_load_print_meta: f_norm_rms_eps   = 0.0e+00
llm_load_print_meta: f_clamp_kqv      = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 8.0e+00
llm_load_print_meta: f_logit_scale    = 0.0e+00
llm_load_print_meta: n_ff             = 3072
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: causal attn      = 0
llm_load_print_meta: pooling type     = 1
llm_load_print_meta: rope type        = -1
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 10000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_ctx_orig_yarn  = 8192
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: ssm_d_conv       = 0
llm_load_print_meta: ssm_d_inner      = 0
llm_load_print_meta: ssm_d_state      = 0
llm_load_print_meta: ssm_dt_rank      = 0
llm_load_print_meta: ssm_dt_b_c_rms   = 0
llm_load_print_meta: model type       = 137M
llm_load_print_meta: model ftype      = Q5_K - Medium
llm_load_print_meta: model params     = 136.78 M
llm_load_print_meta: model size       = 95.16 MiB (5.84 BPW)
llm_load_print_meta: general.name     = Jina Bert Implementation
llm_load_print_meta: UNK token        = 100 '[UNK]'
llm_load_print_meta: SEP token        = 102 '[SEP]'
llm_load_print_meta: PAD token        = 0 '[PAD]'
llm_load_print_meta: CLS token        = 101 '[CLS]'
llm_load_print_meta: MASK token       = 103 '[MASK]'
llm_load_print_meta: LF token         = 0 '[PAD]'
llm_load_print_meta: max token length = 22
ggml_vulkan: Found 1 Vulkan devices:
Vulkan0: AMD Radeon RX 6950 XT (AMD proprietary driver) | uma: 0 | fp16: 1 | warp size: 64
llama_model_load: error loading model: vk::PhysicalDevice::createDevice: ErrorDeviceLost
llama_load_model_from_file: failed to load model
llama_init_from_gpt_params: failed to load model 'models/jina.gguf'
main: error: unable to load model

Name and Version

llama-b3889-bin-win-vulkan-x64

What operating system are you seeing the problem on?

Windows 11

Relevant log output

ggml_vulkan: Found 1 Vulkan devices:
Vulkan0: AMD Radeon RX 6950 XT (AMD proprietary driver) | uma: 0 | fp16: 1 | warp size: 64
llama_model_load: error loading model: vk::PhysicalDevice::createDevice: ErrorDeviceLost
llama_load_model_from_file: failed to load model
llama_init_from_gpt_params: failed to load model 'models/jina.gguf'
main: error: unable to load model