name = "Ornith 1.0 9B"
description = "Open coding-reasoning model for repository tasks and self-improving agents"
family = "ornith"
release_date = "2026-06-25"
last_updated = "2026-06-25"
attachment = true
reasoning = true
temperature = true
tool_call = true
open_weights = true
license = "MIT"

[limit]
context = 262_144

[modalities]
input = ["text", "image"]
output = ["text"]

[[weights]]
label = "Hugging Face"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[links]]
label = "Model card"
url = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"
type = "model_card"

[[links]]
label = "Announcement"
url = "https://deep-reinforce.com/ornith_1_0.html"
type = "announcement"

[[benchmarks]]
name = "SWE-Bench Verified"
score = 69.4
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[benchmarks]]
name = "SWE-Bench Pro"
score = 42.9
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[benchmarks]]
name = "SWE-Bench Multilingual"
score = 52
metric = "percent resolved"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 43.1
metric = "percent"
variant = "Terminus-2"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[benchmarks]]
name = "Terminal-Bench 2.1"
score = 40.6
metric = "percent"
variant = "Claude Code"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[benchmarks]]
name = "NL2Repo"
score = 27.2
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"

[[benchmarks]]
name = "Claw-eval"
score = 63.1
metric = "percent"
source = "https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B"