github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 08:30:05 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #14499] [Bug]: unknown model architecture: 'qwen35moe' when loading Qwen3.5 MoE model in v0.17.4 #55918

New Issue
Closed
opened 2026-04-29 09:56:50 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-29 09:56:50 -05:00
Owner
Copy Link

Originally created by @qwqk423 on GitHub (Feb 27, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14499

What is the issue?

###What happened?
When attempting to generate text using a Qwen 3.5 MoE GGUF model (specifically Qwen3.5 35B A3B Heretic), Ollama fails to load the model and returns an HTTP 500 error. [cite_start]The backend engine throws an error stating that the qwen35moe architecture is unknown[cite: 32].

What did you expect to happen?

Ollama should recognize the qwen35moe architecture and successfully load the model for inference.

Environment Details

  • [cite_start]Ollama Version: 0.17.4 [cite: 1]
  • [cite_start]OS: Windows [cite: 11]
  • [cite_start]GPU: NVIDIA GeForce RTX 5070 Ti (15.9 GiB VRAM) [cite: 3]

Relevant log output

llama_model_loader: - kv   0:                       general.architecture str              = qwen35moe
...
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 19.71 GiB (4.88 BPW) 
llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35moe'
llama_model_load_from_file_impl: failed to load model
time=2026-02-27T20:51:14.940+08:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 error="unable to load model: H:\\ollama_models\\.ollama\\models\\blobs\\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5"
[GIN] 2026/02/27 - 20:51:14 | 500 |    333.4775ms |       127.0.0.1 | POST     "/api/generate"
-----------------------------------------------
------------------------------------------------
ALL:
time=2026-02-27T20:48:41.290+08:00 level=INFO source=routes.go:1663 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:8192 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:H:\\ollama_models\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:true OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[* http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]"
time=2026-02-27T20:48:41.294+08:00 level=INFO source=routes.go:1665 msg="Ollama cloud disabled: true"
time=2026-02-27T20:48:41.294+08:00 level=INFO source=images.go:473 msg="total blobs: 2"
time=2026-02-27T20:48:41.294+08:00 level=INFO source=images.go:480 msg="total unused blobs removed: 0"
time=2026-02-27T20:48:41.294+08:00 level=INFO source=routes.go:1718 msg="Listening on [::]:11434 (version 0.17.4)"
time=2026-02-27T20:48:41.295+08:00 level=INFO source=runner.go:67 msg="discovering available GPUs..."
time=2026-02-27T20:48:41.305+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50982"
time=2026-02-27T20:48:41.442+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50987"
time=2026-02-27T20:48:41.567+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50991"
time=2026-02-27T20:48:41.816+08:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled.  To enable, set OLLAMA_VULKAN=1"
time=2026-02-27T20:48:41.817+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50996"
time=2026-02-27T20:48:41.817+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50997"
time=2026-02-27T20:48:41.817+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50995"
time=2026-02-27T20:48:41.957+08:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-4cef3633-2ee0-08a8-852b-8df403ddb6d1 filter_id="" library=CUDA compute=12.0 name=CUDA0 description="NVIDIA GeForce RTX 5070 Ti" libdirs=ollama,cuda_v13 driver=13.1 pci_id=0000:01:00.0 type=discrete total="15.9 GiB" available="13.5 GiB"
time=2026-02-27T20:48:41.957+08:00 level=INFO source=routes.go:1768 msg="vram-based default context" total_vram="15.9 GiB" default_num_ctx=4096
[GIN] 2026/02/27 - 20:48:41 | 200 |            0s |       127.0.0.1 | GET      "/api/version"
[GIN] 2026/02/27 - 20:48:41 | 200 |      1.2971ms |       127.0.0.1 | GET      "/api/tags"
[GIN] 2026/02/27 - 20:50:15 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/02/27 - 20:50:54 | 201 |   23.8785066s |       127.0.0.1 | POST     "/api/blobs/sha256:499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5"
[GIN] 2026/02/27 - 20:50:54 | 200 |    192.1207ms |       127.0.0.1 | POST     "/api/create"
[GIN] 2026/02/27 - 20:51:14 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/02/27 - 20:51:14 | 200 |     83.4809ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2026/02/27 - 20:51:14 | 200 |     77.4991ms |       127.0.0.1 | POST     "/api/show"
time=2026-02-27T20:51:14.698+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 59523"
time=2026-02-27T20:51:14.809+08:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-27T20:51:14.809+08:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=6 efficiency=0 threads=12
llama_model_loader: loaded meta data with 50 key-value pairs and 733 tensors from H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 (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              = qwen35moe
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                     general.sampling.top_k i32              = 20
llama_model_loader: - kv   3:                     general.sampling.top_p f32              = 0.950000
llama_model_loader: - kv   4:                      general.sampling.temp f32              = 1.000000
llama_model_loader: - kv   5:                               general.name str              = Qwen3.5 35B A3B Heretic
llama_model_loader: - kv   6:                           general.finetune str              = heretic
llama_model_loader: - kv   7:                           general.basename str              = Qwen3.5
llama_model_loader: - kv   8:                         general.size_label str              = 35B-A3B
llama_model_loader: - kv   9:                            general.license str              = apache-2.0
llama_model_loader: - kv  10:                       general.license.link str              = https://huggingface.co/Qwen/Qwen3.5-3...
llama_model_loader: - kv  11:                               general.tags arr[str,5]       = ["heretic", "uncensored", "decensored...
llama_model_loader: - kv  12:                      qwen35moe.block_count u32              = 40
llama_model_loader: - kv  13:                   qwen35moe.context_length u32              = 262144
llama_model_loader: - kv  14:                 qwen35moe.embedding_length u32              = 2048
llama_model_loader: - kv  15:             qwen35moe.attention.head_count u32              = 16
llama_model_loader: - kv  16:          qwen35moe.attention.head_count_kv u32              = 2
llama_model_loader: - kv  17:          qwen35moe.rope.dimension_sections arr[i32,4]       = [11, 11, 10, 0]
llama_model_loader: - kv  18:                   qwen35moe.rope.freq_base f32              = 10000000.000000
llama_model_loader: - kv  19: qwen35moe.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  20:                     qwen35moe.expert_count u32              = 256
llama_model_loader: - kv  21:                qwen35moe.expert_used_count u32              = 8
llama_model_loader: - kv  22:             qwen35moe.attention.key_length u32              = 256
llama_model_loader: - kv  23:           qwen35moe.attention.value_length u32              = 256
llama_model_loader: - kv  24:       qwen35moe.expert_feed_forward_length u32              = 512
llama_model_loader: - kv  25: qwen35moe.expert_shared_feed_forward_length u32              = 512
llama_model_loader: - kv  26:                  qwen35moe.ssm.conv_kernel u32              = 4
llama_model_loader: - kv  27:                   qwen35moe.ssm.state_size u32              = 128
llama_model_loader: - kv  28:                  qwen35moe.ssm.group_count u32              = 16
llama_model_loader: - kv  29:               qwen35moe.ssm.time_step_rank u32              = 32
llama_model_loader: - kv  30:                   qwen35moe.ssm.inner_size u32              = 4096
llama_model_loader: - kv  31:          qwen35moe.full_attention_interval u32              = 4
llama_model_loader: - kv  32:             qwen35moe.rope.dimension_count u32              = 64
llama_model_loader: - kv  33:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  34:                         tokenizer.ggml.pre str              = qwen35
llama_model_loader: - kv  35:                      tokenizer.ggml.tokens arr[str,248320]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  36:                  tokenizer.ggml.token_type arr[i32,248320]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  37:                      tokenizer.ggml.merges arr[str,247587]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  38:                tokenizer.ggml.eos_token_id u32              = 248046
llama_model_loader: - kv  39:            tokenizer.ggml.padding_token_id u32              = 248044
llama_model_loader: - kv  40:                    tokenizer.chat_template str              = {%- set image_count = namespace(value...
llama_model_loader: - kv  41:               general.quantization_version u32              = 2
llama_model_loader: - kv  42:                          general.file_type u32              = 15
llama_model_loader: - kv  43:                                general.url str              = https://huggingface.co/mradermacher/Q...
llama_model_loader: - kv  44:              mradermacher.quantize_version str              = 2
llama_model_loader: - kv  45:                  mradermacher.quantized_by str              = mradermacher
llama_model_loader: - kv  46:                  mradermacher.quantized_at str              = 2026-02-26T08:06:07+01:00
llama_model_loader: - kv  47:                  mradermacher.quantized_on str              = nico1
llama_model_loader: - kv  48:                         general.source.url str              = https://huggingface.co/brayniac/Qwen3...
llama_model_loader: - kv  49:                  mradermacher.convert_type str              = hf
llama_model_loader: - type  f32:  301 tensors
llama_model_loader: - type q4_K:  355 tensors
llama_model_loader: - type q5_K:   30 tensors
llama_model_loader: - type q6_K:   47 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 19.71 GiB (4.88 BPW) 
llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35moe'
llama_model_load_from_file_impl: failed to load model
time=2026-02-27T20:51:14.940+08:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 error="unable to load model: H:\\ollama_models\\.ollama\\models\\blobs\\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5"
[GIN] 2026/02/27 - 20:51:14 | 500 |    333.4775ms |       127.0.0.1 | POST     "/api/generate"
[GIN] 2026/02/27 - 20:52:32 | 200 |            0s |       127.0.0.1 | HEAD     "/"
[GIN] 2026/02/27 - 20:52:32 | 200 |     80.8288ms |       127.0.0.1 | POST     "/api/show"
[GIN] 2026/02/27 - 20:52:32 | 200 |     77.4876ms |       127.0.0.1 | POST     "/api/show"
time=2026-02-27T20:52:32.378+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 61736"
time=2026-02-27T20:52:32.486+08:00 level=INFO source=cpu_windows.go:148 msg=packages count=1
time=2026-02-27T20:52:32.486+08:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=6 efficiency=0 threads=12
llama_model_loader: loaded meta data with 50 key-value pairs and 733 tensors from H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 (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              = qwen35moe
llama_model_loader: - kv   1:                               general.type str              = model
llama_model_loader: - kv   2:                     general.sampling.top_k i32              = 20
llama_model_loader: - kv   3:                     general.sampling.top_p f32              = 0.950000
llama_model_loader: - kv   4:                      general.sampling.temp f32              = 1.000000
llama_model_loader: - kv   5:                               general.name str              = Qwen3.5 35B A3B Heretic
llama_model_loader: - kv   6:                           general.finetune str              = heretic
llama_model_loader: - kv   7:                           general.basename str              = Qwen3.5
llama_model_loader: - kv   8:                         general.size_label str              = 35B-A3B
llama_model_loader: - kv   9:                            general.license str              = apache-2.0
llama_model_loader: - kv  10:                       general.license.link str              = https://huggingface.co/Qwen/Qwen3.5-3...
llama_model_loader: - kv  11:                               general.tags arr[str,5]       = ["heretic", "uncensored", "decensored...
llama_model_loader: - kv  12:                      qwen35moe.block_count u32              = 40
llama_model_loader: - kv  13:                   qwen35moe.context_length u32              = 262144
llama_model_loader: - kv  14:                 qwen35moe.embedding_length u32              = 2048
llama_model_loader: - kv  15:             qwen35moe.attention.head_count u32              = 16
llama_model_loader: - kv  16:          qwen35moe.attention.head_count_kv u32              = 2
llama_model_loader: - kv  17:          qwen35moe.rope.dimension_sections arr[i32,4]       = [11, 11, 10, 0]
llama_model_loader: - kv  18:                   qwen35moe.rope.freq_base f32              = 10000000.000000
llama_model_loader: - kv  19: qwen35moe.attention.layer_norm_rms_epsilon f32              = 0.000001
llama_model_loader: - kv  20:                     qwen35moe.expert_count u32              = 256
llama_model_loader: - kv  21:                qwen35moe.expert_used_count u32              = 8
llama_model_loader: - kv  22:             qwen35moe.attention.key_length u32              = 256
llama_model_loader: - kv  23:           qwen35moe.attention.value_length u32              = 256
llama_model_loader: - kv  24:       qwen35moe.expert_feed_forward_length u32              = 512
llama_model_loader: - kv  25: qwen35moe.expert_shared_feed_forward_length u32              = 512
llama_model_loader: - kv  26:                  qwen35moe.ssm.conv_kernel u32              = 4
llama_model_loader: - kv  27:                   qwen35moe.ssm.state_size u32              = 128
llama_model_loader: - kv  28:                  qwen35moe.ssm.group_count u32              = 16
llama_model_loader: - kv  29:               qwen35moe.ssm.time_step_rank u32              = 32
llama_model_loader: - kv  30:                   qwen35moe.ssm.inner_size u32              = 4096
llama_model_loader: - kv  31:          qwen35moe.full_attention_interval u32              = 4
llama_model_loader: - kv  32:             qwen35moe.rope.dimension_count u32              = 64
llama_model_loader: - kv  33:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  34:                         tokenizer.ggml.pre str              = qwen35
llama_model_loader: - kv  35:                      tokenizer.ggml.tokens arr[str,248320]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  36:                  tokenizer.ggml.token_type arr[i32,248320]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  37:                      tokenizer.ggml.merges arr[str,247587]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
llama_model_loader: - kv  38:                tokenizer.ggml.eos_token_id u32              = 248046
llama_model_loader: - kv  39:            tokenizer.ggml.padding_token_id u32              = 248044
llama_model_loader: - kv  40:                    tokenizer.chat_template str              = {%- set image_count = namespace(value...
llama_model_loader: - kv  41:               general.quantization_version u32              = 2
llama_model_loader: - kv  42:                          general.file_type u32              = 15
llama_model_loader: - kv  43:                                general.url str              = https://huggingface.co/mradermacher/Q...
llama_model_loader: - kv  44:              mradermacher.quantize_version str              = 2
llama_model_loader: - kv  45:                  mradermacher.quantized_by str              = mradermacher
llama_model_loader: - kv  46:                  mradermacher.quantized_at str              = 2026-02-26T08:06:07+01:00
llama_model_loader: - kv  47:                  mradermacher.quantized_on str              = nico1
llama_model_loader: - kv  48:                         general.source.url str              = https://huggingface.co/brayniac/Qwen3...
llama_model_loader: - kv  49:                  mradermacher.convert_type str              = hf
llama_model_loader: - type  f32:  301 tensors
llama_model_loader: - type q4_K:  355 tensors
llama_model_loader: - type q5_K:   30 tensors
llama_model_loader: - type q6_K:   47 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 19.71 GiB (4.88 BPW) 
llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35moe'
llama_model_load_from_file_impl: failed to load model
time=2026-02-27T20:52:32.618+08:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 error="unable to load model: H:\\ollama_models\\.ollama\\models\\blobs\\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5"
[GIN] 2026/02/27 - 20:52:32 | 500 |     331.233ms |       127.0.0.1 | POST     "/api/generate"
[GIN] 2026/02/27 - 20:57:35 | 200 |            0s |       127.0.0.1 | GET      "/api/version"

OS

Windows

GPU

Nvidia

CPU

AMD

Ollama version

0.17.4

Originally created by @qwqk423 on GitHub (Feb 27, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14499 ### What is the issue? ###What happened? When attempting to generate text using a Qwen 3.5 MoE GGUF model (specifically `Qwen3.5 35B A3B Heretic`), Ollama fails to load the model and returns an HTTP 500 error. [cite_start]The backend engine throws an error stating that the `qwen35moe` architecture is unknown[cite: 32]. ### What did you expect to happen? Ollama should recognize the `qwen35moe` architecture and successfully load the model for inference. ### Environment Details * [cite_start]**Ollama Version:** 0.17.4 [cite: 1] * [cite_start]**OS:** Windows [cite: 11] * [cite_start]**GPU:** NVIDIA GeForce RTX 5070 Ti (15.9 GiB VRAM) [cite: 3] ### Relevant log output ```shell llama_model_loader: - kv 0: general.architecture str = qwen35moe ... print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 19.71 GiB (4.88 BPW) llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35moe' llama_model_load_from_file_impl: failed to load model time=2026-02-27T20:51:14.940+08:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 error="unable to load model: H:\\ollama_models\\.ollama\\models\\blobs\\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5" [GIN] 2026/02/27 - 20:51:14 | 500 | 333.4775ms | 127.0.0.1 | POST "/api/generate" ----------------------------------------------- ------------------------------------------------ ALL: time=2026-02-27T20:48:41.290+08:00 level=INFO source=routes.go:1663 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GGML_VK_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:8192 OLLAMA_DEBUG:INFO OLLAMA_EDITOR: OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:H:\\ollama_models\\.ollama\\models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NO_CLOUD:true OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[* http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_REMOTES:[ollama.com] OLLAMA_SCHED_SPREAD:false OLLAMA_VULKAN:false ROCR_VISIBLE_DEVICES:]" time=2026-02-27T20:48:41.294+08:00 level=INFO source=routes.go:1665 msg="Ollama cloud disabled: true" time=2026-02-27T20:48:41.294+08:00 level=INFO source=images.go:473 msg="total blobs: 2" time=2026-02-27T20:48:41.294+08:00 level=INFO source=images.go:480 msg="total unused blobs removed: 0" time=2026-02-27T20:48:41.294+08:00 level=INFO source=routes.go:1718 msg="Listening on [::]:11434 (version 0.17.4)" time=2026-02-27T20:48:41.295+08:00 level=INFO source=runner.go:67 msg="discovering available GPUs..." time=2026-02-27T20:48:41.305+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50982" time=2026-02-27T20:48:41.442+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50987" time=2026-02-27T20:48:41.567+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50991" time=2026-02-27T20:48:41.816+08:00 level=INFO source=runner.go:106 msg="experimental Vulkan support disabled. To enable, set OLLAMA_VULKAN=1" time=2026-02-27T20:48:41.817+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50996" time=2026-02-27T20:48:41.817+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50997" time=2026-02-27T20:48:41.817+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 50995" time=2026-02-27T20:48:41.957+08:00 level=INFO source=types.go:42 msg="inference compute" id=GPU-4cef3633-2ee0-08a8-852b-8df403ddb6d1 filter_id="" library=CUDA compute=12.0 name=CUDA0 description="NVIDIA GeForce RTX 5070 Ti" libdirs=ollama,cuda_v13 driver=13.1 pci_id=0000:01:00.0 type=discrete total="15.9 GiB" available="13.5 GiB" time=2026-02-27T20:48:41.957+08:00 level=INFO source=routes.go:1768 msg="vram-based default context" total_vram="15.9 GiB" default_num_ctx=4096 [GIN] 2026/02/27 - 20:48:41 | 200 | 0s | 127.0.0.1 | GET "/api/version" [GIN] 2026/02/27 - 20:48:41 | 200 | 1.2971ms | 127.0.0.1 | GET "/api/tags" [GIN] 2026/02/27 - 20:50:15 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/27 - 20:50:54 | 201 | 23.8785066s | 127.0.0.1 | POST "/api/blobs/sha256:499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5" [GIN] 2026/02/27 - 20:50:54 | 200 | 192.1207ms | 127.0.0.1 | POST "/api/create" [GIN] 2026/02/27 - 20:51:14 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/27 - 20:51:14 | 200 | 83.4809ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/27 - 20:51:14 | 200 | 77.4991ms | 127.0.0.1 | POST "/api/show" time=2026-02-27T20:51:14.698+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 59523" time=2026-02-27T20:51:14.809+08:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-27T20:51:14.809+08:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=6 efficiency=0 threads=12 llama_model_loader: loaded meta data with 50 key-value pairs and 733 tensors from H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 (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 = qwen35moe llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.sampling.top_k i32 = 20 llama_model_loader: - kv 3: general.sampling.top_p f32 = 0.950000 llama_model_loader: - kv 4: general.sampling.temp f32 = 1.000000 llama_model_loader: - kv 5: general.name str = Qwen3.5 35B A3B Heretic llama_model_loader: - kv 6: general.finetune str = heretic llama_model_loader: - kv 7: general.basename str = Qwen3.5 llama_model_loader: - kv 8: general.size_label str = 35B-A3B llama_model_loader: - kv 9: general.license str = apache-2.0 llama_model_loader: - kv 10: general.license.link str = https://huggingface.co/Qwen/Qwen3.5-3... llama_model_loader: - kv 11: general.tags arr[str,5] = ["heretic", "uncensored", "decensored... llama_model_loader: - kv 12: qwen35moe.block_count u32 = 40 llama_model_loader: - kv 13: qwen35moe.context_length u32 = 262144 llama_model_loader: - kv 14: qwen35moe.embedding_length u32 = 2048 llama_model_loader: - kv 15: qwen35moe.attention.head_count u32 = 16 llama_model_loader: - kv 16: qwen35moe.attention.head_count_kv u32 = 2 llama_model_loader: - kv 17: qwen35moe.rope.dimension_sections arr[i32,4] = [11, 11, 10, 0] llama_model_loader: - kv 18: qwen35moe.rope.freq_base f32 = 10000000.000000 llama_model_loader: - kv 19: qwen35moe.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 20: qwen35moe.expert_count u32 = 256 llama_model_loader: - kv 21: qwen35moe.expert_used_count u32 = 8 llama_model_loader: - kv 22: qwen35moe.attention.key_length u32 = 256 llama_model_loader: - kv 23: qwen35moe.attention.value_length u32 = 256 llama_model_loader: - kv 24: qwen35moe.expert_feed_forward_length u32 = 512 llama_model_loader: - kv 25: qwen35moe.expert_shared_feed_forward_length u32 = 512 llama_model_loader: - kv 26: qwen35moe.ssm.conv_kernel u32 = 4 llama_model_loader: - kv 27: qwen35moe.ssm.state_size u32 = 128 llama_model_loader: - kv 28: qwen35moe.ssm.group_count u32 = 16 llama_model_loader: - kv 29: qwen35moe.ssm.time_step_rank u32 = 32 llama_model_loader: - kv 30: qwen35moe.ssm.inner_size u32 = 4096 llama_model_loader: - kv 31: qwen35moe.full_attention_interval u32 = 4 llama_model_loader: - kv 32: qwen35moe.rope.dimension_count u32 = 64 llama_model_loader: - kv 33: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 34: tokenizer.ggml.pre str = qwen35 llama_model_loader: - kv 35: tokenizer.ggml.tokens arr[str,248320] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 36: tokenizer.ggml.token_type arr[i32,248320] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 37: tokenizer.ggml.merges arr[str,247587] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 38: tokenizer.ggml.eos_token_id u32 = 248046 llama_model_loader: - kv 39: tokenizer.ggml.padding_token_id u32 = 248044 llama_model_loader: - kv 40: tokenizer.chat_template str = {%- set image_count = namespace(value... llama_model_loader: - kv 41: general.quantization_version u32 = 2 llama_model_loader: - kv 42: general.file_type u32 = 15 llama_model_loader: - kv 43: general.url str = https://huggingface.co/mradermacher/Q... llama_model_loader: - kv 44: mradermacher.quantize_version str = 2 llama_model_loader: - kv 45: mradermacher.quantized_by str = mradermacher llama_model_loader: - kv 46: mradermacher.quantized_at str = 2026-02-26T08:06:07+01:00 llama_model_loader: - kv 47: mradermacher.quantized_on str = nico1 llama_model_loader: - kv 48: general.source.url str = https://huggingface.co/brayniac/Qwen3... llama_model_loader: - kv 49: mradermacher.convert_type str = hf llama_model_loader: - type f32: 301 tensors llama_model_loader: - type q4_K: 355 tensors llama_model_loader: - type q5_K: 30 tensors llama_model_loader: - type q6_K: 47 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 19.71 GiB (4.88 BPW) llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35moe' llama_model_load_from_file_impl: failed to load model time=2026-02-27T20:51:14.940+08:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 error="unable to load model: H:\\ollama_models\\.ollama\\models\\blobs\\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5" [GIN] 2026/02/27 - 20:51:14 | 500 | 333.4775ms | 127.0.0.1 | POST "/api/generate" [GIN] 2026/02/27 - 20:52:32 | 200 | 0s | 127.0.0.1 | HEAD "/" [GIN] 2026/02/27 - 20:52:32 | 200 | 80.8288ms | 127.0.0.1 | POST "/api/show" [GIN] 2026/02/27 - 20:52:32 | 200 | 77.4876ms | 127.0.0.1 | POST "/api/show" time=2026-02-27T20:52:32.378+08:00 level=INFO source=server.go:431 msg="starting runner" cmd="C:\\Users\\33108\\AppData\\Local\\Programs\\Ollama\\ollama.exe runner --ollama-engine --port 61736" time=2026-02-27T20:52:32.486+08:00 level=INFO source=cpu_windows.go:148 msg=packages count=1 time=2026-02-27T20:52:32.486+08:00 level=INFO source=cpu_windows.go:195 msg="" package=0 cores=6 efficiency=0 threads=12 llama_model_loader: loaded meta data with 50 key-value pairs and 733 tensors from H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 (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 = qwen35moe llama_model_loader: - kv 1: general.type str = model llama_model_loader: - kv 2: general.sampling.top_k i32 = 20 llama_model_loader: - kv 3: general.sampling.top_p f32 = 0.950000 llama_model_loader: - kv 4: general.sampling.temp f32 = 1.000000 llama_model_loader: - kv 5: general.name str = Qwen3.5 35B A3B Heretic llama_model_loader: - kv 6: general.finetune str = heretic llama_model_loader: - kv 7: general.basename str = Qwen3.5 llama_model_loader: - kv 8: general.size_label str = 35B-A3B llama_model_loader: - kv 9: general.license str = apache-2.0 llama_model_loader: - kv 10: general.license.link str = https://huggingface.co/Qwen/Qwen3.5-3... llama_model_loader: - kv 11: general.tags arr[str,5] = ["heretic", "uncensored", "decensored... llama_model_loader: - kv 12: qwen35moe.block_count u32 = 40 llama_model_loader: - kv 13: qwen35moe.context_length u32 = 262144 llama_model_loader: - kv 14: qwen35moe.embedding_length u32 = 2048 llama_model_loader: - kv 15: qwen35moe.attention.head_count u32 = 16 llama_model_loader: - kv 16: qwen35moe.attention.head_count_kv u32 = 2 llama_model_loader: - kv 17: qwen35moe.rope.dimension_sections arr[i32,4] = [11, 11, 10, 0] llama_model_loader: - kv 18: qwen35moe.rope.freq_base f32 = 10000000.000000 llama_model_loader: - kv 19: qwen35moe.attention.layer_norm_rms_epsilon f32 = 0.000001 llama_model_loader: - kv 20: qwen35moe.expert_count u32 = 256 llama_model_loader: - kv 21: qwen35moe.expert_used_count u32 = 8 llama_model_loader: - kv 22: qwen35moe.attention.key_length u32 = 256 llama_model_loader: - kv 23: qwen35moe.attention.value_length u32 = 256 llama_model_loader: - kv 24: qwen35moe.expert_feed_forward_length u32 = 512 llama_model_loader: - kv 25: qwen35moe.expert_shared_feed_forward_length u32 = 512 llama_model_loader: - kv 26: qwen35moe.ssm.conv_kernel u32 = 4 llama_model_loader: - kv 27: qwen35moe.ssm.state_size u32 = 128 llama_model_loader: - kv 28: qwen35moe.ssm.group_count u32 = 16 llama_model_loader: - kv 29: qwen35moe.ssm.time_step_rank u32 = 32 llama_model_loader: - kv 30: qwen35moe.ssm.inner_size u32 = 4096 llama_model_loader: - kv 31: qwen35moe.full_attention_interval u32 = 4 llama_model_loader: - kv 32: qwen35moe.rope.dimension_count u32 = 64 llama_model_loader: - kv 33: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 34: tokenizer.ggml.pre str = qwen35 llama_model_loader: - kv 35: tokenizer.ggml.tokens arr[str,248320] = ["!", "\"", "#", "$", "%", "&", "'", ... llama_model_loader: - kv 36: tokenizer.ggml.token_type arr[i32,248320] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 37: tokenizer.ggml.merges arr[str,247587] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... llama_model_loader: - kv 38: tokenizer.ggml.eos_token_id u32 = 248046 llama_model_loader: - kv 39: tokenizer.ggml.padding_token_id u32 = 248044 llama_model_loader: - kv 40: tokenizer.chat_template str = {%- set image_count = namespace(value... llama_model_loader: - kv 41: general.quantization_version u32 = 2 llama_model_loader: - kv 42: general.file_type u32 = 15 llama_model_loader: - kv 43: general.url str = https://huggingface.co/mradermacher/Q... llama_model_loader: - kv 44: mradermacher.quantize_version str = 2 llama_model_loader: - kv 45: mradermacher.quantized_by str = mradermacher llama_model_loader: - kv 46: mradermacher.quantized_at str = 2026-02-26T08:06:07+01:00 llama_model_loader: - kv 47: mradermacher.quantized_on str = nico1 llama_model_loader: - kv 48: general.source.url str = https://huggingface.co/brayniac/Qwen3... llama_model_loader: - kv 49: mradermacher.convert_type str = hf llama_model_loader: - type f32: 301 tensors llama_model_loader: - type q4_K: 355 tensors llama_model_loader: - type q5_K: 30 tensors llama_model_loader: - type q6_K: 47 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 19.71 GiB (4.88 BPW) llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35moe' llama_model_load_from_file_impl: failed to load model time=2026-02-27T20:52:32.618+08:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=H:\ollama_models\.ollama\models\blobs\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5 error="unable to load model: H:\\ollama_models\\.ollama\\models\\blobs\\sha256-499c2c0a0394da8c0c7c22a93850d75e743579f1320d4d056f1db28c2045aba5" [GIN] 2026/02/27 - 20:52:32 | 500 | 331.233ms | 127.0.0.1 | POST "/api/generate" [GIN] 2026/02/27 - 20:57:35 | 200 | 0s | 127.0.0.1 | GET "/api/version" ``` ### OS Windows ### GPU Nvidia ### CPU AMD ### Ollama version 0.17.4
GiteaMirror added the bug label 2026-04-29 09:56:50 -05:00
GiteaMirror commented 2026-04-29 09:56:52 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 27, 2026):

qwen3.5 models from HF are not currently supported in ollama, needs either a vendor sync (#14134) or tweaking of the go runner.

<!-- gh-comment-id:3972880207 --> @rick-github commented on GitHub (Feb 27, 2026): qwen3.5 models from HF are not currently supported in ollama, needs either a vendor sync (#14134) or tweaking of the go runner.
GiteaMirror commented 2026-04-29 09:56:53 -05:00
Author
Owner
Copy Link

@qwqk423 commented on GitHub (Feb 27, 2026):

Okay, got it, thank you

Hugging Face 上的 qwen3.5 模型目前尚未在 Ollama 中得到支持，需要进行供应商同步（ #14134 ）或调整 Go 运行器。

<!-- gh-comment-id:3972938674 --> @qwqk423 commented on GitHub (Feb 27, 2026): Okay, got it, thank you > Hugging Face 上的 qwen3.5 模型目前尚未在 Ollama 中得到支持，需要进行供应商同步（ [#14134](https://github.com/ollama/ollama/pull/14134) ）或调整 Go 运行器。
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#55918
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?