github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #14513] Unable to load Qwen3.5-27B #35177

New Issue
Closed
opened 2026-04-22 19:30:06 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-22 19:30:06 -05:00
Owner
Copy Link

Originally created by @giveen on GitHub (Feb 28, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/14513

What is the issue?

When trying to load Qwen3.5-27B I get a 500 error. I have ensured I am running ollama 0.17.4 and have restarted the service to ensure that its properly loaded. I have no other issues with other models.

jabbatheduck@404notfound:~$ ollama run qwen3.5-27b-Q8_0:latest
Error: 500 Internal Server Error: unable to load model: /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655

jabbatheduck@404notfound:~$ ollama ls qwen3.5-27b-Q8_0:latest
NAME                       ID              SIZE     MODIFIED   
qwen3.5-27b-Q8_0:latest    e5c3170404ad    28 GB    2 days ago

Model was done with a HF->gguf

convert_hf_to_gguf.py /home/jabbatheduck/.cache/huggingface/hub/models--Qwen--Qwen3.5-27B/snapshots/b7ca741b86de18df552fd2cc952861e04621a4bd \
  --outfile Qwen3.5-27B-Q8_0.gguf \
  --outtype q8_0

Relevant log output

28 09:40:26 404notfound ollama[1938551]: llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35'
Feb 28 09:40:26 404notfound ollama[1938551]: llama_model_load_from_file_impl: failed to load model
Feb 28 09:40:26 404notfound ollama[1938551]: time=2026-02-28T09:40:26.327-07:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=/mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 error="unable to load model: /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655"

Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: loaded meta data with 39 key-value pairs and 851 tensors from /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 (version GGUF V3 (latest))
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   0:                       general.architecture str              = qwen35
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   1:                               general.type str              = model
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   2:                     general.sampling.top_k i32              = 20
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   3:                     general.sampling.top_p f32              = 0.950000
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   4:                      general.sampling.temp f32              = 0.600000
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   5:                               general.name str              = B7Ca741B86De18Df552Fd2Cc952861E04621A4Bd
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   6:                         general.size_label str              = 27B
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   7:                            general.license str              = apache-2.0
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   8:                       general.license.link str              = https://huggingface.co/Qwen/Qwen3.5-2...
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv   9:                               general.tags arr[str,1]       = ["image-text-to-text"]
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  10:                         qwen35.block_count u32              = 64
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  11:                      qwen35.context_length u32              = 262144
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  12:                    qwen35.embedding_length u32              = 5120
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  13:                 qwen35.feed_forward_length u32              = 17408
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  14:                qwen35.attention.head_count u32              = 24
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  15:             qwen35.attention.head_count_kv u32              = 4
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  16:             qwen35.rope.dimension_sections arr[i32,4]       = [11, 11, 10, 0]
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  17:                      qwen35.rope.freq_base f32              = 10000000.000000
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  18:    qwen35.attention.layer_norm_rms_epsilon f32              = 0.000001
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  19:                qwen35.attention.key_length u32              = 256
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  20:              qwen35.attention.value_length u32              = 256
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  21:                          general.file_type u32              = 7
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  22:                     qwen35.ssm.conv_kernel u32              = 4
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  23:                      qwen35.ssm.state_size u32              = 128
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  24:                     qwen35.ssm.group_count u32              = 16
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  25:                  qwen35.ssm.time_step_rank u32              = 48
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  26:                      qwen35.ssm.inner_size u32              = 6144
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  27:             qwen35.full_attention_interval u32              = 4
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  28:                qwen35.rope.dimension_count u32              = 64
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  29:               general.quantization_version u32              = 2
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  30:                       tokenizer.ggml.model str              = gpt2
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  31:                         tokenizer.ggml.pre str              = qwen35
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  32:                      tokenizer.ggml.tokens arr[str,248320]  = ["!", "\"", "#", "$", "%", "&", "'", ...
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  33:                  tokenizer.ggml.token_type arr[i32,248320]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  34:                      tokenizer.ggml.merges arr[str,247587]  = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",...
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  35:                tokenizer.ggml.eos_token_id u32              = 248046
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  36:            tokenizer.ggml.padding_token_id u32              = 248044
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  37:               tokenizer.ggml.add_bos_token bool             = false
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv  38:                    tokenizer.chat_template str              = {%- set image_count = namespace(value...
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - type  f32:  353 tensors
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - type q8_0:  498 tensors
Feb 28 09:41:56 404notfound ollama[1938551]: print_info: file format = GGUF V3 (latest)
Feb 28 09:41:56 404notfound ollama[1938551]: print_info: file type   = Q8_0
Feb 28 09:41:56 404notfound ollama[1938551]: print_info: file size   = 26.62 GiB (8.50 BPW)
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35'
Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_load_from_file_impl: failed to load model
Feb 28 09:41:56 404notfound ollama[1938551]: time=2026-02-28T09:41:56.293-07:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=/mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 error="unable to load model: /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655"
Feb 28 09:41:56 404notfound ollama[1938551]: [GIN] 2026/02/28 - 09:41:56 | 500 |  355.214767ms |       127.0.0.1 | POST     "/api/generate"
Feb 28 09:42:48 404notfound ollama[1938551]: [GIN] 2026/02/28 - 09:42:48 | 200 |      30.844µs |       127.0.0.1 | GET      "/api/version"

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.17.4

Originally created by @giveen on GitHub (Feb 28, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/14513 ### What is the issue? When trying to load Qwen3.5-27B I get a 500 error. I have ensured I am running ollama 0.17.4 and have restarted the service to ensure that its properly loaded. I have no other issues with other models. ```bash jabbatheduck@404notfound:~$ ollama run qwen3.5-27b-Q8_0:latest Error: 500 Internal Server Error: unable to load model: /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 ``` ```bash jabbatheduck@404notfound:~$ ollama ls qwen3.5-27b-Q8_0:latest NAME ID SIZE MODIFIED qwen3.5-27b-Q8_0:latest e5c3170404ad 28 GB 2 days ago ``` Model was done with a HF->gguf ```python convert_hf_to_gguf.py /home/jabbatheduck/.cache/huggingface/hub/models--Qwen--Qwen3.5-27B/snapshots/b7ca741b86de18df552fd2cc952861e04621a4bd \ --outfile Qwen3.5-27B-Q8_0.gguf \ --outtype q8_0 ``` ### Relevant log output ```shell 28 09:40:26 404notfound ollama[1938551]: llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35' Feb 28 09:40:26 404notfound ollama[1938551]: llama_model_load_from_file_impl: failed to load model Feb 28 09:40:26 404notfound ollama[1938551]: time=2026-02-28T09:40:26.327-07:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=/mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 error="unable to load model: /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655" ``` ```bash Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: loaded meta data with 39 key-value pairs and 851 tensors from /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 (version GGUF V3 (latest)) Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 0: general.architecture str = qwen35 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 1: general.type str = model Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 2: general.sampling.top_k i32 = 20 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 3: general.sampling.top_p f32 = 0.950000 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 4: general.sampling.temp f32 = 0.600000 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 5: general.name str = B7Ca741B86De18Df552Fd2Cc952861E04621A4Bd Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 6: general.size_label str = 27B Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 7: general.license str = apache-2.0 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 8: general.license.link str = https://huggingface.co/Qwen/Qwen3.5-2... Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 9: general.tags arr[str,1] = ["image-text-to-text"] Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 10: qwen35.block_count u32 = 64 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 11: qwen35.context_length u32 = 262144 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 12: qwen35.embedding_length u32 = 5120 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 13: qwen35.feed_forward_length u32 = 17408 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 14: qwen35.attention.head_count u32 = 24 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 15: qwen35.attention.head_count_kv u32 = 4 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 16: qwen35.rope.dimension_sections arr[i32,4] = [11, 11, 10, 0] Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 17: qwen35.rope.freq_base f32 = 10000000.000000 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 18: qwen35.attention.layer_norm_rms_epsilon f32 = 0.000001 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 19: qwen35.attention.key_length u32 = 256 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 20: qwen35.attention.value_length u32 = 256 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 21: general.file_type u32 = 7 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 22: qwen35.ssm.conv_kernel u32 = 4 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 23: qwen35.ssm.state_size u32 = 128 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 24: qwen35.ssm.group_count u32 = 16 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 25: qwen35.ssm.time_step_rank u32 = 48 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 26: qwen35.ssm.inner_size u32 = 6144 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 27: qwen35.full_attention_interval u32 = 4 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 28: qwen35.rope.dimension_count u32 = 64 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 29: general.quantization_version u32 = 2 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 30: tokenizer.ggml.model str = gpt2 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 31: tokenizer.ggml.pre str = qwen35 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 32: tokenizer.ggml.tokens arr[str,248320] = ["!", "\"", "#", "$", "%", "&", "'", ... Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 33: tokenizer.ggml.token_type arr[i32,248320] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 34: tokenizer.ggml.merges arr[str,247587] = ["Ġ Ġ", "ĠĠ ĠĠ", "i n", "Ġ t",... Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 35: tokenizer.ggml.eos_token_id u32 = 248046 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 36: tokenizer.ggml.padding_token_id u32 = 248044 Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 37: tokenizer.ggml.add_bos_token bool = false Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - kv 38: tokenizer.chat_template str = {%- set image_count = namespace(value... Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - type f32: 353 tensors Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_loader: - type q8_0: 498 tensors Feb 28 09:41:56 404notfound ollama[1938551]: print_info: file format = GGUF V3 (latest) Feb 28 09:41:56 404notfound ollama[1938551]: print_info: file type = Q8_0 Feb 28 09:41:56 404notfound ollama[1938551]: print_info: file size = 26.62 GiB (8.50 BPW) Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'qwen35' Feb 28 09:41:56 404notfound ollama[1938551]: llama_model_load_from_file_impl: failed to load model Feb 28 09:41:56 404notfound ollama[1938551]: time=2026-02-28T09:41:56.293-07:00 level=INFO source=sched.go:473 msg="NewLlamaServer failed" model=/mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655 error="unable to load model: /mnt/storage/Ollama/models/blobs/sha256-2f45cee0cd2bfd3ce6e30606612106cd10f2b7dc8f9f01e3521958869ec16655" Feb 28 09:41:56 404notfound ollama[1938551]: [GIN] 2026/02/28 - 09:41:56 | 500 | 355.214767ms | 127.0.0.1 | POST "/api/generate" Feb 28 09:42:48 404notfound ollama[1938551]: [GIN] 2026/02/28 - 09:42:48 | 200 | 30.844µs | 127.0.0.1 | GET "/api/version" ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.17.4
GiteaMirror added the bug label 2026-04-22 19:30:06 -05:00
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#35177
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?