github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #13177] deepseek-ocr :: Not able to run this model #70772

New Issue
Closed
opened 2026-05-04 22:56:45 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-05-04 22:56:45 -05:00
Owner
Copy Link

Originally created by @amey6992 on GitHub (Nov 20, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/13177

What is the issue?

Issue with deepseek-ocr

  • I did pull the model
  • When I run it, I get an error
  • Error: 500 Internal Server Error: unable to load model: /home/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c
Image

free memory is about 160Gbs, the other model llama3.2 worked well

Relevant log output

llama_model_loader: loaded meta data with 33 key-value pairs and 631 tensors from /home/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c (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:           deepseekocr.attention.head_count u32              = 10
llama_model_loader: - kv   1:        deepseekocr.attention.head_count_kv u32              = 10
llama_model_loader: - kv   2:                    deepseekocr.block_count u32              = 12
llama_model_loader: - kv   3:                 deepseekocr.context_length u32              = 8192
llama_model_loader: - kv   4:               deepseekocr.embedding_length u32              = 1280
llama_model_loader: - kv   5:                   deepseekocr.expert_count u32              = 64
llama_model_loader: - kv   6:              deepseekocr.expert_used_count u32              = 6
llama_model_loader: - kv   7:            deepseekocr.feed_forward_length u32              = 6848
llama_model_loader: - kv   8:                       general.architecture str              = deepseekocr
llama_model_loader: - kv   9:                          general.file_type u32              = 1
llama_model_loader: - kv  10:               general.quantization_version u32              = 2
llama_model_loader: - kv  11:      deepseekocr.leading_dense_block_count u32              = 1
llama_model_loader: - kv  12:                deepseekocr.sam.block_count u32              = 12
llama_model_loader: - kv  13:           deepseekocr.sam.embedding_length u32              = 768
llama_model_loader: - kv  14:   deepseekocr.sam.global_attention_indexes arr[i32,4]       = [2, 5, 8, 11]
llama_model_loader: - kv  15:                 deepseekocr.sam.head_count u32              = 12
llama_model_loader: - kv  16:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  17:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  18:           tokenizer.ggml.add_padding_token bool             = false
llama_model_loader: - kv  19:                tokenizer.ggml.bos_token_id u32              = 0
llama_model_loader: - kv  20:                tokenizer.ggml.eos_token_id u32              = 1
llama_model_loader: - kv  21:                      tokenizer.ggml.merges arr[str,127741]  = ["Ġ t", "Ġ a", "i n", "Ġ Ġ", "h e...
llama_model_loader: - kv  22:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  23:            tokenizer.ggml.padding_token_id u32              = 2
llama_model_loader: - kv  24:                         tokenizer.ggml.pre str              = default
llama_model_loader: - kv  25:                      tokenizer.ggml.scores arr[f32,129280]  = [0.000000, 1.000000, 2.000000, 3.0000...
llama_model_loader: - kv  26:                  tokenizer.ggml.token_type arr[i32,129280]  = [3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  27:                      tokenizer.ggml.tokens arr[str,129280]  = ["<｜begin▁of▁sentence｜>", "<�...
llama_model_loader: - kv  28:             deepseekocr.vision.block_count u32              = 24
llama_model_loader: - kv  29:        deepseekocr.vision.embedding_length u32              = 1024
llama_model_loader: - kv  30:              deepseekocr.vision.head_count u32              = 16
llama_model_loader: - kv  31:              deepseekocr.vision.image_size u32              = 224
llama_model_loader: - kv  32:              deepseekocr.vision.patch_size u32              = 14
llama_model_loader: - type  f32:  360 tensors
llama_model_loader: - type  f16:  151 tensors
llama_model_loader: - type bf16:  120 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = F16
print_info: file size   = 6.22 GiB (16.02 BPW) 
llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'deepseekocr'
llama_model_load_from_file_impl: failed to load model
time=2025-11-21T01:23:08.526+05:30 level=INFO source=sched.go:425 msg="NewLlamaServer failed" model=/home/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c error="unable to load model: /home/plus91/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c"

OS

Linux

GPU

No response

CPU

Intel

Ollama version

0.12.11

Originally created by @amey6992 on GitHub (Nov 20, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/13177 ### What is the issue? ### Issue with deepseek-ocr ### - I did pull the model - When I run it, I get an error - Error: 500 Internal Server Error: unable to load model: /home/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c <img width="1854" height="847" alt="Image" src="https://github.com/user-attachments/assets/e16dcd79-215a-46ec-8058-fedc6a3fd09f" /> free memory is about 160Gbs, the other model llama3.2 worked well ### Relevant log output ```shell llama_model_loader: loaded meta data with 33 key-value pairs and 631 tensors from /home/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c (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: deepseekocr.attention.head_count u32 = 10 llama_model_loader: - kv 1: deepseekocr.attention.head_count_kv u32 = 10 llama_model_loader: - kv 2: deepseekocr.block_count u32 = 12 llama_model_loader: - kv 3: deepseekocr.context_length u32 = 8192 llama_model_loader: - kv 4: deepseekocr.embedding_length u32 = 1280 llama_model_loader: - kv 5: deepseekocr.expert_count u32 = 64 llama_model_loader: - kv 6: deepseekocr.expert_used_count u32 = 6 llama_model_loader: - kv 7: deepseekocr.feed_forward_length u32 = 6848 llama_model_loader: - kv 8: general.architecture str = deepseekocr llama_model_loader: - kv 9: general.file_type u32 = 1 llama_model_loader: - kv 10: general.quantization_version u32 = 2 llama_model_loader: - kv 11: deepseekocr.leading_dense_block_count u32 = 1 llama_model_loader: - kv 12: deepseekocr.sam.block_count u32 = 12 llama_model_loader: - kv 13: deepseekocr.sam.embedding_length u32 = 768 llama_model_loader: - kv 14: deepseekocr.sam.global_attention_indexes arr[i32,4] = [2, 5, 8, 11] llama_model_loader: - kv 15: deepseekocr.sam.head_count u32 = 12 llama_model_loader: - kv 16: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 17: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 18: tokenizer.ggml.add_padding_token bool = false llama_model_loader: - kv 19: tokenizer.ggml.bos_token_id u32 = 0 llama_model_loader: - kv 20: tokenizer.ggml.eos_token_id u32 = 1 llama_model_loader: - kv 21: tokenizer.ggml.merges arr[str,127741] = ["Ġ t", "Ġ a", "i n", "Ġ Ġ", "h e... llama_model_loader: - kv 22: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 23: tokenizer.ggml.padding_token_id u32 = 2 llama_model_loader: - kv 24: tokenizer.ggml.pre str = default llama_model_loader: - kv 25: tokenizer.ggml.scores arr[f32,129280] = [0.000000, 1.000000, 2.000000, 3.0000... llama_model_loader: - kv 26: tokenizer.ggml.token_type arr[i32,129280] = [3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... llama_model_loader: - kv 27: tokenizer.ggml.tokens arr[str,129280] = ["<｜begin▁of▁sentence｜>", "<�... llama_model_loader: - kv 28: deepseekocr.vision.block_count u32 = 24 llama_model_loader: - kv 29: deepseekocr.vision.embedding_length u32 = 1024 llama_model_loader: - kv 30: deepseekocr.vision.head_count u32 = 16 llama_model_loader: - kv 31: deepseekocr.vision.image_size u32 = 224 llama_model_loader: - kv 32: deepseekocr.vision.patch_size u32 = 14 llama_model_loader: - type f32: 360 tensors llama_model_loader: - type f16: 151 tensors llama_model_loader: - type bf16: 120 tensors print_info: file format = GGUF V3 (latest) print_info: file type = F16 print_info: file size = 6.22 GiB (16.02 BPW) llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'deepseekocr' llama_model_load_from_file_impl: failed to load model time=2025-11-21T01:23:08.526+05:30 level=INFO source=sched.go:425 msg="NewLlamaServer failed" model=/home/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c error="unable to load model: /home/plus91/.ollama/models/blobs/sha256-3a18673ff291a1d8de94d490877127899356d33a18028d5f3945bf245c11b02c" ``` ### OS Linux ### GPU _No response_ ### CPU Intel ### Ollama version 0.12.11
GiteaMirror added the bug label 2026-05-04 22:56:45 -05:00
GiteaMirror commented 2026-05-04 22:56:53 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 20, 2025):

DeepSeek-OCR requires Ollama v0.13.0 or later.

<!-- gh-comment-id:3559907380 --> @rick-github commented on GitHub (Nov 20, 2025): DeepSeek-OCR requires [Ollama v0.13.0](https://github.com/ollama/ollama/releases) or later.
GiteaMirror commented 2026-05-04 22:56:55 -05:00
Author
Owner
Copy Link

@amey6992 commented on GitHub (Nov 20, 2025):

Yes, just got this from the docs

<!-- gh-comment-id:3559913526 --> @amey6992 commented on GitHub (Nov 20, 2025): Yes, just got this from the docs
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#70772
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?