"CUDA error: invalid device function" with Ollama on Windows with AMD 780M

[bryndin]

Downloaded and installed AMD-Software-PRO-Edition-23.Q4-Win10-Win11-For-HIP.exe

Replaced "c:\Program Files\AMD\ROCm\5.7\bin\rocblas.dll" and "c:\Program Files\AMD\ROCm\5.7\bin\rocblas\library"

with the content in rocm gfx1103 AMD780M phoenix V3.7z

Ollama (Windows) can find the GPU, but then fails with

ggml_cuda_compute_forward: RMS_NORM failed CUDA error: invalid device function current device: 0, in function ggml_cuda_compute_forward at C:/a/ollama/ollama/llm/llama.cpp/ggml-cuda.cu:2300 err GGML_ASSERT: C:/a/ollama/ollama/llm/llama.cpp/ggml-cuda.cu:60: !"CUDA error"

Any idea what could be wrong? I tried all other versions of gfx1103 files you have.

Win 11, latest Ollama with llama3:8b-instruct-q6_K AMD drivers: amd-software-adrenalin-edition-24.4.1-minimalsetup-240423_web.exe

Here's the full log

time=2024-05-07T03:15:51.386-07:00 level=INFO source=images.go:828 msg="total blobs: 5"
time=2024-05-07T03:15:51.386-07:00 level=INFO source=images.go:835 msg="total unused blobs removed: 0"
time=2024-05-07T03:15:51.387-07:00 level=INFO source=routes.go:1071 msg="Listening on 127.0.0.1:11434 (version 0.1.33)"
time=2024-05-07T03:15:51.387-07:00 level=INFO source=payload.go:44 msg="Dynamic LLM libraries [cpu cpu_avx cpu_avx2 cuda_v11.3 rocm_v5.7]"
time=2024-05-07T03:15:51.387-07:00 level=INFO source=gpu.go:96 msg="Detecting GPUs"
time=2024-05-07T03:15:51.391-07:00 level=INFO source=cpu_common.go:11 msg="CPU has AVX2"
time=2024-05-07T03:15:51.405-07:00 level=INFO source=amd_windows.go:39 msg="AMD Driver: 50732000"
time=2024-05-07T03:15:51.415-07:00 level=INFO source=amd_windows.go:68 msg="detected hip devices" count=1
time=2024-05-07T03:15:51.415-07:00 level=INFO source=amd_windows.go:88 msg="hip device" id=0 name="AMD Radeon(TM) 780M" gfx=gfx1103
time=2024-05-07T03:15:51.415-07:00 level=INFO source=amd_windows.go:109 msg="amdgpu is supported" gpu=0 gpu_type=gfx1103
time=2024-05-07T03:15:51.671-07:00 level=INFO source=amd_windows.go:127 msg="amdgpu memory" gpu=0 total="29375.0 MiB"
time=2024-05-07T03:15:51.671-07:00 level=INFO source=amd_windows.go:128 msg="amdgpu memory" gpu=0 available="29230.9 MiB"
[GIN] 2024/05/07 - 03:15:51 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2024/05/07 - 03:15:51 | 200 |      1.6367ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2024/05/07 - 03:15:51 | 200 |      1.5727ms |       127.0.0.1 | POST     "/api/show"
time=2024-05-07T03:15:51.780-07:00 level=INFO source=gpu.go:96 msg="Detecting GPUs"
time=2024-05-07T03:15:51.785-07:00 level=INFO source=cpu_common.go:11 msg="CPU has AVX2"
time=2024-05-07T03:15:51.793-07:00 level=INFO source=amd_windows.go:39 msg="AMD Driver: 50732000"
time=2024-05-07T03:15:51.794-07:00 level=INFO source=amd_windows.go:68 msg="detected hip devices" count=1
time=2024-05-07T03:15:51.794-07:00 level=INFO source=amd_windows.go:88 msg="hip device" id=0 name="AMD Radeon(TM) 780M" gfx=gfx1103
time=2024-05-07T03:15:51.794-07:00 level=INFO source=amd_windows.go:109 msg="amdgpu is supported" gpu=0 gpu_type=gfx1103
time=2024-05-07T03:15:52.044-07:00 level=INFO source=amd_windows.go:127 msg="amdgpu memory" gpu=0 total="29375.0 MiB"
time=2024-05-07T03:15:52.049-07:00 level=INFO source=amd_windows.go:128 msg="amdgpu memory" gpu=0 available="29230.9 MiB"
time=2024-05-07T03:15:54.013-07:00 level=INFO source=memory.go:152 msg="offload to gpu" layers.real=-1 layers.estimate=33 memory.available="29230.9 MiB" memory.required.full="6749.0 MiB" memory.required.partial="6749.0 MiB" memory.required.kv="256.0 MiB" memory.weights.total="5872.0 MiB" memory.weights.repeating="5461.0 MiB" memory.weights.nonrepeating="411.0 MiB" memory.graph.full="164.0 MiB" memory.graph.partial="677.5 MiB"
time=2024-05-07T03:15:54.014-07:00 level=INFO source=memory.go:152 msg="offload to gpu" layers.real=-1 layers.estimate=33 memory.available="29230.9 MiB" memory.required.full="6749.0 MiB" memory.required.partial="6749.0 MiB" memory.required.kv="256.0 MiB" memory.weights.total="5872.0 MiB" memory.weights.repeating="5461.0 MiB" memory.weights.nonrepeating="411.0 MiB" memory.graph.full="164.0 MiB" memory.graph.partial="677.5 MiB"
time=2024-05-07T03:15:54.014-07:00 level=INFO source=cpu_common.go:11 msg="CPU has AVX2"
time=2024-05-07T03:15:54.023-07:00 level=INFO source=server.go:289 msg="starting llama server" cmd="C:\\Users\\brynd\\AppData\\Local\\Programs\\Ollama\\ollama_runners\\rocm_v5.7\\ollama_llama_server.exe --model C:\\Users\\brynd\\.ollama\\models\\blobs\\sha256-3f75702e9f27f9b481928f2df8e2c011ae9b27e18821edc48ce5953160a2cb93 --ctx-size 2048 --batch-size 512 --embedding --log-disable --n-gpu-layers 33 --parallel 1 --port 50928"
time=2024-05-07T03:15:54.030-07:00 level=INFO source=sched.go:340 msg="loaded runners" count=1
time=2024-05-07T03:15:54.030-07:00 level=INFO source=server.go:432 msg="waiting for llama runner to start responding"
{"function":"server_params_parse","level":"INFO","line":2606,"msg":"logging to file is disabled.","tid":"3604","timestamp":1715076954}
{"build":2770,"commit":"952d03d","function":"wmain","level":"INFO","line":2823,"msg":"build info","tid":"3604","timestamp":1715076954}
{"function":"wmain","level":"INFO","line":2830,"msg":"system info","n_threads":8,"n_threads_batch":-1,"system_info":"AVX = 1 | AVX_VNNI = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 | ","tid":"3604","timestamp":1715076954,"total_threads":16}
llama_model_loader: loaded meta data with 21 key-value pairs and 291 tensors from C:\Users\brynd\.ollama\models\blobs\sha256-3f75702e9f27f9b481928f2df8e2c011ae9b27e18821edc48ce5953160a2cb93 (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              = Meta-Llama-3-8B-Instruct
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              = 18
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.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  15:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  16:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  17:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  18:                tokenizer.ggml.eos_token_id u32              = 128001
llama_model_loader: - kv  19:                    tokenizer.chat_template str              = {% set loop_messages = messages %}{% ...
llama_model_loader: - kv  20:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   65 tensors
llama_model_loader: - type q6_K:  226 tensors
llm_load_vocab: missing pre-tokenizer type, using: 'default'
llm_load_vocab:                                             
llm_load_vocab: ************************************        
llm_load_vocab: GENERATION QUALITY WILL BE DEGRADED!        
llm_load_vocab: CONSIDER REGENERATING THE MODEL             
llm_load_vocab: ************************************        
llm_load_vocab:                                             
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: f_logit_scale    = 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: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 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: 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: model type       = 8B
llm_load_print_meta: model ftype      = Q6_K
llm_load_print_meta: model params     = 8.03 B
llm_load_print_meta: model size       = 6.14 GiB (6.56 BPW) 
llm_load_print_meta: general.name     = Meta-Llama-3-8B-Instruct
llm_load_print_meta: BOS token        = 128000 '<|begin_of_text|>'
llm_load_print_meta: EOS token        = 128001 '<|end_of_text|>'
llm_load_print_meta: LF token         = 128 'Ä'
llm_load_print_meta: EOT token        = 128009 '<|eot_id|>'
ggml_cuda_init: GGML_CUDA_FORCE_MMQ:   no
ggml_cuda_init: CUDA_USE_TENSOR_CORES: yes
ggml_cuda_init: found 1 ROCm devices:
  Device 0: AMD Radeon(TM) 780M, compute capability 11.0, VMM: no
llm_load_tensors: ggml ctx size =    0.30 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:      ROCm0 buffer size =  5871.99 MiB
llm_load_tensors:        CPU buffer size =   410.98 MiB
.........................................................................................
llama_new_context_with_model: n_ctx      = 2048
llama_new_context_with_model: n_batch    = 512
llama_new_context_with_model: n_ubatch   = 512
llama_new_context_with_model: freq_base  = 500000.0
llama_new_context_with_model: freq_scale = 1
llama_kv_cache_init:      ROCm0 KV buffer size =   256.00 MiB
llama_new_context_with_model: KV self size  =  256.00 MiB, K (f16):  128.00 MiB, V (f16):  128.00 MiB
llama_new_context_with_model:  ROCm_Host  output buffer size =     0.50 MiB
llama_new_context_with_model:      ROCm0 compute buffer size =   258.50 MiB
llama_new_context_with_model:  ROCm_Host compute buffer size =    12.01 MiB
llama_new_context_with_model: graph nodes  = 1030
llama_new_context_with_model: graph splits = 2
ggml_cuda_compute_forward: RMS_NORM failed
CUDA error: invalid device function
  current device: 0, in function ggml_cuda_compute_forward at C:/a/ollama/ollama/llm/llama.cpp/ggml-cuda.cu:2300
  err
GGML_ASSERT: C:/a/ollama/ollama/llm/llama.cpp/ggml-cuda.cu:60: !"CUDA error"

[likelovewant]

Did you get your ollama here ? (https://github.com/likelovewant/ollama-for-amd/releases/tag/V0.1.33-alpha) That's seems the problem of cuda , if you have any other Navida gpu connect your pc ,if so that's the issue came from . otherwise you are using something like zluda, try to remove the nvml.dll from the zluda folder . and restar the ollama again . if not working , try to follow this guide to build your own (https://github.com/likelovewant/ollama-for-amd/blob/main/wiki%20for%20build%20on%20amd]((ollama))

[cpz2006]

I used ollama v0.1.30(windows)，installed hips and zluda，use command:""D:\Zluda\zluda.exe "C:\Users\……\AppData\Local\Programs\Ollama\ollama app.exe"""manually，could run ollama on 780m successfully.But when I updated ollama，had the same error"ggml_cuda_compute_forward: RMS_NORM failed".

[likelovewant]

I used ollama v0.1.30(windows)，installed hips and zluda，use command:""D:\Zluda\zluda.exe "C:\Users\……\AppData\Local\Programs\Ollama\ollama app.exe"""manually，could run ollama on 780m successfully.But when I updated ollama，had the same error"ggml_cuda_compute_forward: RMS_NORM failed".

sorry ,this a not repo for ollama , if you have ollama issue ,you may raise an isuse there https://github.com/likelovewant/ollama-for-amd, rocmlibs for gfx1103 should not be the issue. anyway , if you download the ollama installer from above rellease page, That's because I did not not install any cuda tool kit during the build wich is for nvida gpu. In the past few version of ollama ,ollama has several bugs will wrong take amd zluda as nvida . and zluda has fixed it .but still not perferct . Try to remove the the nvml.dll from the zluda folder . and restar the ollama again . only use rocm rather than zluda . otherwise , you may try to build your version of ollama. It's works pefect on my pc .

[bryndin]

Did you get your ollama here ? (https://github.com/likelovewant/ollama-for-amd/releases/tag/V0.1.33-alpha)

Replacing the original Ollama with your files in ollama-windows-amd64.zip did work. Thank you! I wonder why the same can't be integrated into Ollama itself.

[likelovewant]

Did you get your ollama here ? (https://github.com/likelovewant/ollama-for-amd/releases/tag/V0.1.33-alpha)

Replacing the original Ollama with your files in ollama-windows-amd64.zip did work. Thank you! I wonder why the same can't be integrated into Ollama itself.

Try the ollama instller ,it's natively support for amd780m ,you don't need to replace any other file .

[cpz2006]

Thanks,this ollama (https://github.com/likelovewant/ollama-for-amd/releases/tag/V0.1.33-alpha) works on my 780m apu，I can use the tiny vision language model moondream on ollama now.

OK,other large language model and vision language model llava worked,but moondream couldn't.

[likelovewant]

Thanks,this ollama (https://github.com/likelovewant/ollama-for-amd/releases/tag/V0.1.33-alpha) works on my 780m apu，I can use the tiny vision language model moondream on ollama now.

OK,other large language model and vision language model llava worked,but moondream couldn't.

Try the new update vesrionv0.1.34-alpha or git clone ollama ,edit as guide on wiki for build on amd [ollama-for-amd],build your own ollama

[cpz2006]

Thanks,this ollama (https://github.com/likelovewant/ollama-for-amd/releases/tag/V0.1.33-alpha) works on my 780m apu，I can use the tiny vision language model moondream on ollama now. OK,other large language model and vision language model llava worked,but moondream couldn't.

Try the new update vesrionv0.1.34-alpha or git clone ollama ,edit as guide on wiki for build on amd [ollama-for-amd],build your own ollama

I download from https://github.com/likelovewant/ollama-for-amd/releases/download/v0.1.34-alpha/ollama-windows-amd64.zip and overwrite ollama,but couldn't use model moondream. use command:ollama -v return:ollama version is 0.1.33-alpha-5-g3952ceb some error in server.log: llama_model_loader: loaded meta data with 20 key-value pairs and 245 tensors from C:\Users\^.ollama\models\blobs\sha256-e554c6b9de016673fd2c732e0342967727e9659ca5f853a4947cc96263fa602b (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 = phi2 llama_model_loader: - kv 1: general.name str = moondream2 llama_model_loader: - kv 2: phi2.context_length u32 = 2048 llama_model_loader: - kv 3: phi2.embedding_length u32 = 2048 llama_model_loader: - kv 4: phi2.feed_forward_length u32 = 8192 llama_model_loader: - kv 5: phi2.block_count u32 = 24 llama_model_loader: - kv 6: phi2.attention.head_count u32 = 32 llama_model_loader: - kv 7: phi2.attention.head_count_kv u32 = 32 llama_model_loader: - kv 8: phi2.attention.layer_norm_epsilon f32 = 0.000010 llama_model_loader: - kv 9: phi2.rope.dimension_count u32 = 32 llama_model_loader: - kv 10: general.file_type u32 = 2 llama_model_loader: - kv 11: tokenizer.ggml.add_bos_token bool = false llama_model_loader: - kv 12: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,51200] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 14: tokenizer.ggml.token_type arr[i32,51200] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 15: tokenizer.ggml.merges arr[str,50000] = ["Ġ t", "Ġ a", "h e", "i n", "r e",... llama_model_loader: - kv 16: tokenizer.ggml.bos_token_id u32 = 50256 llama_model_loader: - kv 17: tokenizer.ggml.eos_token_id u32 = 50256 llama_model_loader: - kv 18: tokenizer.ggml.unknown_token_id u32 = 50256 llama_model_loader: - kv 19: general.quantization_version u32 = 2 llama_model_loader: - type f32: 147 tensors llama_model_loader: - type q4_0: 97 tensors llama_model_loader: - type q6_K: 1 tensors llm_load_vocab: missing pre-tokenizer type, using: 'default' llm_load_vocab:
llm_load_vocab: ****
llm_load_vocab: GENERATION QUALITY WILL BE DEGRADED!
llm_load_vocab: CONSIDER REGENERATING THE MODEL
llm_load_vocab: ****
llm_load_vocab:
llm_load_vocab: mismatch in special tokens definition ( 910/51200 vs 944/51200 ). llm_load_print_meta: format = GGUF V3 (latest) llm_load_print_meta: arch = phi2 llm_load_print_meta: vocab type = BPE llm_load_print_meta: n_vocab = 51200 llm_load_print_meta: n_merges = 50000 llm_load_print_meta: n_ctx_train = 2048 llm_load_print_meta: n_embd = 2048 llm_load_print_meta: n_head = 32 llm_load_print_meta: n_head_kv = 32 llm_load_print_meta: n_layer = 24 llm_load_print_meta: n_rot = 32 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 = 2048 llm_load_print_meta: n_embd_v_gqa = 2048 llm_load_print_meta: f_norm_eps = 1.0e-05 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 = 0.0e+00 llm_load_print_meta: f_logit_scale = 0.0e+00 llm_load_print_meta: n_ff = 8192 llm_load_print_meta: n_expert = 0 llm_load_print_meta: n_expert_used = 0 llm_load_print_meta: causal attn = 1 llm_load_print_meta: pooling type = 0 llm_load_print_meta: rope type = 2 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_yarn_orig_ctx = 2048 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: model type = 1B llm_load_print_meta: model ftype = Q4_0 llm_load_print_meta: model params = 1.42 B llm_load_print_meta: model size = 788.55 MiB (4.66 BPW) llm_load_print_meta: general.name = moondream2 llm_load_print_meta: BOS token = 50256 '<|endoftext|>' llm_load_print_meta: EOS token = 50256 '<|endoftext|>' llm_load_print_meta: UNK token = 50256 '<|endoftext|>' llm_load_print_meta: LF token = 128 'Ä' llm_load_tensors: ggml ctx size = 0.24 MiB llm_load_tensors: offloading 24 repeating layers to GPU llm_load_tensors: offloading non-repeating layers to GPU llm_load_tensors: offloaded 25/25 layers to GPU llm_load_tensors: ROCm0 buffer size = 732.30 MiB llm_load_tensors: CPU buffer size = 56.25 MiB .............................................................................. llama_new_context_with_model: n_ctx = 2048 llama_new_context_with_model: n_batch = 512 llama_new_context_with_model: n_ubatch = 512 llama_new_context_with_model: freq_base = 10000.0 llama_new_context_with_model: freq_scale = 1 llama_kv_cache_init: ROCm0 KV buffer size = 384.00 MiB llama_new_context_with_model: KV self size = 384.00 MiB, K (f16): 192.00 MiB, V (f16): 192.00 MiB llama_new_context_with_model: ROCm_Host output buffer size = 0.20 MiB llama_new_context_with_model: ROCm0 compute buffer size = 160.00 MiB llama_new_context_with_model: ROCm_Host compute buffer size = 8.01 MiB llama_new_context_with_model: graph nodes = 921 llama_new_context_with_model: graph splits = 2 ggml_cuda_compute_forward: MUL_MAT failed CUDA error: named symbol not found current device: 0, in function ggml_cuda_compute_forward at D:/ollama-for-amd/llm/llama.cpp/ggml-cuda.cu:2300 err GGML_ASSERT: D:/ollama-for-amd/llm/llama.cpp/ggml-cuda.cu:60: !"CUDA error"

[likelovewant]

0.1.33-alpha-5 The 0.1.33-alpha-5 is exactly same as v0.1.34 ollama offocial , as my repo was aim show ollama can support more amd gpu ,so there is no offoicial release over there . llama_new_context_with_model: KV self size = 384.00 MiB, K (f16): 192.00 MiB, V (f16): 192.00 MiB llama_new_context_with_model: ROCm_Host output buffer size = 0.20 MiB llama_new_context_with_model: ROCm0 compute buffer size = 160.00 MiB llama_new_context_with_model: ROCm_Host compute buffer size = 8.01 MiB

The ROCm is working ollama is working but something not override ollama offocial release which not from my repo,that's the cause of cuda error.

make sure fully close the ollama .and reinstall the OllamaSetup.exe rather the [ollama-windows-amd64.zip]

also make sure the nvdml.dll and nvcuda.dll not in the zluda folder.

if not try to remove the moodream model and download again .