]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
model : update label for LFM2-24B-A2B (#19848)
authorTarek Dakhran <redacted>
Tue, 24 Feb 2026 13:27:42 +0000 (14:27 +0100)
committerGitHub <redacted>
Tue, 24 Feb 2026 13:27:42 +0000 (14:27 +0100)
commitda426cb25031928bcbc0d822bbd5ac3491ed4c13
tree51f460b24b034753ec2c8935675135bfebb5f1e4
parentc830f99cfa79d7e627e48de32280838f97b41115
model : update label for LFM2-24B-A2B (#19848)

* model : Update label for LFM2-24B-A2B

```
❯ build/bin/llama-bench -m /data/playground/checkpoints/LFM2-24B-A2B-Preview-Q4_0.gguf,/data/playground/checkpoints/LFM2-8B-A1B-Q4_0.gguf -p 1 -n 0
| model                          |       size |     params | backend    | threads |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | ------: | --------------: | -------------------: |
| lfm2moe 24B.A2B Q4_0           |  12.54 GiB |    23.84 B | CPU        |      10 |             pp1 |         30.35 ± 2.49 |
| lfm2moe 8B.A1B Q4_0            |   4.41 GiB |     8.34 B | CPU        |      10 |             pp1 |         49.24 ± 1.93 |
```

* Remove extra line
src/llama-model.cpp
src/llama-model.h