# Error: pull model manifest: file does not exist: /usr/share/ollama/.ollama/models/blobs/sha256-...

- **ID:** `llm/ollama-model-not-found-pull-failed`
- **Domain:** llm
- **Category:** install_error
- **Verification:** ai_generated
- **Fix Rate:** 88%

## Root Cause

Ollama model pull fails because a required blob file is missing from the local model cache, often due to incomplete download, disk corruption, or permission issues in the model storage directory.

## Version Compatibility

| Version | Status | Introduced | Deprecated |
|---------|--------|------------|------------|
| Ollama 0.1.20 | active | — | — |
| Ollama 0.1.25 | active | — | — |
| Ollama 0.1.30 | active | — | — |
| Linux amd64 | active | — | — |
| macOS arm64 | active | — | — |

## Workarounds

1. **Remove the corrupted model and re-pull it cleanly:

ollama rm llama2
rm -rf ~/.ollama/models/blobs/sha256-*
ollama pull llama2** (90% success)
   ```
   Remove the corrupted model and re-pull it cleanly:

ollama rm llama2
rm -rf ~/.ollama/models/blobs/sha256-*
ollama pull llama2
   ```
2. **Check disk space and permissions, then restart Ollama service before retrying:

df -h /usr/share/ollama/
sudo chown -R ollama:ollama /usr/share/ollama/.ollama
sudo systemctl restart ollama
ollama pull llama2** (85% success)
   ```
   Check disk space and permissions, then restart Ollama service before retrying:

df -h /usr/share/ollama/
sudo chown -R ollama:ollama /usr/share/ollama/.ollama
sudo systemctl restart ollama
ollama pull llama2
   ```

## Dead Ends

- **** — Creates an empty file that fails integrity checks; Ollama will still error on hash verification. (100% fail)
- **** — Changes file ownership to root, causing subsequent permission errors for non-root Ollama processes. (50% fail)