feat(gallery): add Ornith 1.5 35B variants (#11625)

* feat(gallery): add Ornith 1.5 35B variants

Add the supported APEX, MTP, and official Q8_0 builds for the multimodal Ornith 1.5 35B model.

Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* fix(gallery): restore Ornith 1.5 entries

The 35B insertion split the existing 9B entry and nested its remaining
fields under the final 35B file. Restore the 9B fields before the new
35B entries so the gallery index parses correctly.

Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>
This commit is contained in:
authored and GitHub committed 2026-09-11 15:26:50 +02:00
1 parent b845468a11
commit 1e5b8aeca5
1 file changed
+211
+211
View File
@@ -2104,6 +2104,217 @@
- filename: llama-cpp/mmproj/ornith-1.5-9b/mmproj-BF16.gguf
uri: huggingface://ornith-ai/Ornith-1.5-9B-GGUF/mmproj-Ornith-1.5-9B-BF16.gguf
sha256: 626f9f90627402a6bf4a999111d0fbd69b5fcca7aa8ba089d69e5f10e8858e1d
- &ornith-1-5-35b-a3b-apex
name: "ornith-1.5-35b-a3b-apex"
variants:
- model: ornith-1.5-35b-a3b-apex-compact
- model: ornith-1.5-35b-a3b-mtp-apex
- model: ornith-1.5-35b-a3b-mtp-apex-compact
url: "github:mudler/LocalAI/gallery/virtual.yaml@master"
urls:
- https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B
- https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF
- https://huggingface.co/mudler/Ornith-1.5-35B-A3B-APEX-GGUF
- https://huggingface.co/mudler/Ornith-1.5-35B-A3B-APEX-MTP-GGUF
description: |
Ornith-1.5-35B-A3B is an MIT-licensed Qwen3.5 mixture-of-experts model
from Ornith AI for agentic coding, reasoning, repository-level software
tasks, and tool use. It supports text and image input with a context window
of 262K tokens.
This default entry uses the APEX Balanced GGUF and BF16 vision projector.
Compact APEX and MTP-enabled APEX builds are available as variants.
license: "mit"
tags:
- llm
- gguf
- cpu
- gpu
- qwen
- reasoning
- thinking
- coding
- agent
- tools
- vision
- multimodal
- moe
- long-context
last_checked: "2026-08-20"
overrides:
backend: llama-cpp
context_size: 262144
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
options:
- use_jinja:true
parameters:
min_p: 0
model: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-Balanced.gguf
repeat_penalty: 1
temperature: 0.6
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
files:
- filename: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-Balanced.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-GGUF/Ornith-1.5-35B-A3B-APEX-Balanced.gguf
sha256: b86ab34c9c767974f44e2624e6f7409c0bc96100fd96f3a97b4d12c1f3c9c223
- filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-GGUF/mmproj.gguf
sha256: e277123529451fbcff6d880ad5bd7ada999f9d5e6339ba3639d06c5cef4763b1
- !!merge <<: *ornith-1-5-35b-a3b-apex
name: "ornith-1.5-35b-a3b-apex-compact"
variants: []
description: |
Ornith-1.5-35B-A3B in the smaller APEX Compact GGUF format, with the
shared BF16 vision projector for multimodal prompts.
overrides:
backend: llama-cpp
context_size: 262144
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
options:
- use_jinja:true
parameters:
min_p: 0
model: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-Compact.gguf
repeat_penalty: 1
temperature: 0.6
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
files:
- filename: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-Compact.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-GGUF/Ornith-1.5-35B-A3B-APEX-Compact.gguf
sha256: 846eb4121c1b28df0e2dff06c3f3d174084231a7400c649ba02023843ea41021
- filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-GGUF/mmproj.gguf
sha256: e277123529451fbcff6d880ad5bd7ada999f9d5e6339ba3639d06c5cef4763b1
- !!merge <<: *ornith-1-5-35b-a3b-apex
name: "ornith-1.5-35b-a3b-mtp-apex"
variants: []
description: |
Ornith-1.5-35B-A3B in the APEX Balanced GGUF format with native
multi-token prediction enabled for speculative decoding, plus the shared
BF16 vision projector.
tags:
- llm
- gguf
- cpu
- gpu
- qwen
- reasoning
- thinking
- coding
- agent
- tools
- vision
- multimodal
- moe
- long-context
- mtp
overrides:
backend: llama-cpp
context_size: 262144
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
options:
- use_jinja:true
- spec_type:draft-mtp
- spec_n_max:6
- spec_p_min:0.75
parameters:
min_p: 0
model: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-MTP-Balanced.gguf
repeat_penalty: 1
temperature: 0.6
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
files:
- filename: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-MTP-Balanced.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-MTP-GGUF/Ornith-1.5-35B-A3B-APEX-MTP-Balanced.gguf
sha256: 964fbf455e4ecb73499341688134abc1513882690de61e5e2f383724888e18d3
- filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-MTP-GGUF/mmproj.gguf
sha256: e277123529451fbcff6d880ad5bd7ada999f9d5e6339ba3639d06c5cef4763b1
- !!merge <<: *ornith-1-5-35b-a3b-apex
name: "ornith-1.5-35b-a3b-mtp-apex-compact"
variants: []
description: |
Ornith-1.5-35B-A3B in the APEX Compact GGUF format with native multi-token
prediction enabled for speculative decoding, plus the shared BF16 vision
projector.
tags:
- llm
- gguf
- cpu
- gpu
- qwen
- reasoning
- thinking
- coding
- agent
- tools
- vision
- multimodal
- moe
- long-context
- mtp
overrides:
backend: llama-cpp
context_size: 262144
function:
automatic_tool_parsing_fallback: true
grammar:
disable: true
known_usecases:
- chat
- vision
mmproj: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
options:
- use_jinja:true
- spec_type:draft-mtp
- spec_n_max:6
- spec_p_min:0.75
parameters:
min_p: 0
model: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-MTP-Compact.gguf
repeat_penalty: 1
temperature: 0.6
top_k: 20
top_p: 0.95
template:
use_tokenizer_template: true
files:
- filename: llama-cpp/models/ornith-1.5-35b-a3b/Ornith-1.5-35B-A3B-APEX-MTP-Compact.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-MTP-GGUF/Ornith-1.5-35B-A3B-APEX-MTP-Compact.gguf
sha256: a7373f724cf3693e8a6a5f8209ae9a06ff8071ae2af60659bdbdf8ff6f6c10df
- filename: llama-cpp/mmproj/ornith-1.5-35b-a3b/mmproj.gguf
uri: huggingface://mudler/Ornith-1.5-35B-A3B-APEX-MTP-GGUF/mmproj.gguf
sha256: e277123529451fbcff6d880ad5bd7ada999f9d5e6339ba3639d06c5cef4763b1
- !!merge <<: *ornith-1-5-9b
name: "ornith-1.5-9b-q8"
variants: []