frob/glm-ocr:bf16

frob/ glm-ocr:bf16

7 Downloads Updated 2 days ago

Fix missing metadata.

vision tools

ollama run frob/glm-ocr:bf16

curl http://localhost:11434/api/chat \
  -d '{
    "model": "frob/glm-ocr:bf16",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='frob/glm-ocr:bf16',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'frob/glm-ocr:bf16',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 2 days ago

2 days ago

899cbcb09e30 · 2.2GB

model

archglmocr

·

parameters1.11B

·

quantizationF16

2.2GB

params

{ "temperature": 0 }

18B

Readme

$ diff -u <(dumpgguf --no-tensor $(ollama show --modelfile glm-ocr:bf16 | grep -oP "^FROM \K\S+")) <(dumpgguf --no-tensor $(ollama show --modelfile frob/glm-ocr:bf16 | grep -oP "^FROM \K\S+"))
--- /dev/fd/63	2026-10-06 01:20:54.578479396 +0200
+++ /dev/fd/62	2026-10-06 01:20:54.578479396 +0200
@@ -1,9 +1,9 @@
-INFO:gguf-dump:* Loading: /root/.ollama/models/blobs/sha256-65493e1f85b9ea4ba3ed793515fde13cbdbea7d74ad2c662b566b146eab0081e
+INFO:gguf-dump:* Loading: /root/.ollama/models/blobs/sha256-b2891fe9ce386254e6ad799ff3a03d2b262f22f21973549ca1e71b71260817be
 * File is LITTLE endian, script is running on a LITTLE endian host.
-* Dumping 49 key/value pair(s)
+* Dumping 50 key/value pair(s)
       1: UINT32     |        1 | GGUF.version = 3
       2: UINT64     |        1 | GGUF.tensor_count = 527
-      3: UINT64     |        1 | GGUF.kv_count = 46
+      3: UINT64     |        1 | GGUF.kv_count = 47
       4: STRING     |        1 | general.architecture = 'glmocr'
       5: UINT32     |        1 | general.file_type = 1
       6: UINT32     |        1 | general.quantization_version = 2
@@ -50,3 +50,4 @@
      47: [FLOAT32]  |    59392 | tokenizer.ggml.scores = [0.0, 1.0, 2.0, 3.0, 4.0, 5.0, ...]
      48: [INT32]    |    59392 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...]
      49: [STRING]   |    59392 | tokenizer.ggml.tokens = ['Ā', 'ā', 'Ă', 'ă', 'Ą', 'ą', ...]
+     50: UINT32     |        1 | tokenizer.ggml.eot_token_id = 59253

```console
$ diff -u <(dumpgguf --no-tensor $(ollama show --modelfile glm-ocr:bf16 | grep -oP "^FROM \K\S+")) <(dumpgguf --no-tensor $(ollama show --modelfile frob/glm-ocr:bf16 | grep -oP "^FROM \K\S+"))
--- /dev/fd/63	2026-10-06 01:20:54.578479396 +0200
+++ /dev/fd/62	2026-10-06 01:20:54.578479396 +0200
@@ -1,9 +1,9 @@
-INFO:gguf-dump:* Loading: /root/.ollama/models/blobs/sha256-65493e1f85b9ea4ba3ed793515fde13cbdbea7d74ad2c662b566b146eab0081e
+INFO:gguf-dump:* Loading: /root/.ollama/models/blobs/sha256-b2891fe9ce386254e6ad799ff3a03d2b262f22f21973549ca1e71b71260817be
 * File is LITTLE endian, script is running on a LITTLE endian host.
-* Dumping 49 key/value pair(s)
+* Dumping 50 key/value pair(s)
       1: UINT32     |        1 | GGUF.version = 3
       2: UINT64     |        1 | GGUF.tensor_count = 527
-      3: UINT64     |        1 | GGUF.kv_count = 46
+      3: UINT64     |        1 | GGUF.kv_count = 47
       4: STRING     |        1 | general.architecture = 'glmocr'
       5: UINT32     |        1 | general.file_type = 1
       6: UINT32     |        1 | general.quantization_version = 2
@@ -50,3 +50,4 @@
      47: [FLOAT32]  |    59392 | tokenizer.ggml.scores = [0.0, 1.0, 2.0, 3.0, 4.0, 5.0, ...]
      48: [INT32]    |    59392 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...]
      49: [STRING]   |    59392 | tokenizer.ggml.tokens = ['Ā', 'ā', 'Ă', 'ă', 'Ą', 'ą', ...]
+     50: UINT32     |        1 | tokenizer.ggml.eot_token_id = 59253
```

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)