whatmodel.sh / Ornith 1.5 35B-A3B

Ornith 1.5 35B-A3B

+ contribute

Ornith 1.5 35B-A3B is an open-weights language model from DeepReinforce, released August 2026. Its published size is 35.95B parameters (3B active per token) and a 262k-token context window. Licence: mit.

Catalogue identity notes

DeepReinforce's Ornith 1.5 35B-A3B (ornith-ai/Ornith-1.5-35B-A3B @10fbf86, created 2026-08-18, MIT). Lab rows from its own card table, which also prints the same-lab Ornith 1.0 35B column (anchored to other makers through Kwaipilot's independent Ornith-1.0-35B run).

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 35,951,822,704 parameters; active 3B from the maker's 'A3B' naming. ollama: 35.5B LM + 447M vision projector.

70.6±1.5#78 of 719 LLMsfrom 12 benchmark results across 1 source

$ how it compares

The LLMs ranked either side of Ornith 1.5 35B-A3B, out of 828 with a score, then more from DeepReinforce.

#	model	score	where	RAM	released
81	GLM-5.1 Z.ai	71±1	api $2.15/M +local	~119GB	2026-04
82	Qwen3.5 397B-A17B Alibaba	71±0	api $1.09/M +local	~65GB	2026-02
83	Ornith 1.5 35B-A3B DeepReinforce	71±2	local	~22GB	2026-08
84	MiniMax-M3 MiniMax	70±1	api $0.48/M +local	~137GB	2026-06
85	Grok 4.3 SpaceXAI (xAI)	70±1	api $1.56/M		2026-04
	more from DeepReinforce
122	Ornith 1.0 35B (35B-A3B) DeepReinforce	68±1	local	~23GB	2026-06

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #67 of 300 · 68% of its score · 9 tests

Beats Gemini 3.6 Flash, GPT-5 Pro, LongCat 2.0 (1.6T-A48B) here, though they rank higher overall. Trails MiMo-V2.5 (310B-A15B), MiniMax-M3 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Pro (split unspecified) lab	59.6% ✎	▃·▁▁▁▂▂▃█▇▃▁ beats 40/55	very hard · avg 10%	▇▇▇▇▇ 13%	sharp · at its level · small run
SWE-bench Verified lab	79.0% ✎	▁▁▁▁▁▁▂▄█▆▁▁ beats 123/131	hard · avg 26%	▇▇▇▇ 11%	sharp · small run
Terminal-Bench 2.1 lab	67.8% ✎	▁·▁·▁▁▄▃▄█▃▇ beats 33/56	hard · avg 18%	▇▇▇▇ 10%	at its level · small run
SWE-bench Multilingual lab	71.4% ✎	▁··▁▃▂▅▇█▃▃▁ beats 36/60	hard · avg 28%	▇▇▇ 7.3%	at its level · small run
BrowseComp lab	67.6% ✎	▂·▄▁▂▃▆▄█▂▃▄ beats 30/42	hard · avg 23%	▇▇▇ 7.0%	at its level · small run
MCP-Atlas (public set) lab	70.2% ✎	▂·▁▂▁·▁▃█▆▆▂ beats 25/37	hard · avg 29%	▇▇▇ 6.5%	at its level · small run
Humanity's Last Exam with tools lab	33.4% ✎	▂▂▂▁▄▂▃▆█▇▂▆ beats 12/54	very hard · avg 13%	▇▇ 5.4%	at its level · small run
WideSearch lab	67.8% ✎	█·▂▆··▂···▂▂ beats 6/10	medium · avg 35%	▇▇ 4.5%	at its level · small run

1 more test

test	result	vs the others in its run	difficulty	counts	why
Toolathlon-Verified lab	48.7% ✎	▂·▂▂▅▃···▂██ beats 4/18	hard · avg 27%	▇ 3.5%	blunt · at its level · small run

coding▁▁▁▃▆██▇▆▄▃▁ #61 of 458 · 31% of its score · 3 tests

Beats DeepSeek V4 Flash Vision (Exp), Gemini 3.1 Pro (Preview), GPT-5.2 here, though they rank higher overall. Trails MiMo-V2.5-Pro (1T-A42B), MiniMax-M3 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Pro (split unspecified) lab	59.6% ✎	▃·▁▁▁▂▂▃█▇▃▁ beats 40/55	very hard · avg 10%	▇▇▇▇▇ 13%	sharp · at its level · small run
SWE-bench Verified lab	79.0% ✎	▁▁▁▁▁▁▂▄█▆▁▁ beats 123/131	hard · avg 26%	▇▇▇▇ 11%	sharp · small run
SWE-bench Multilingual lab	71.4% ✎	▁··▁▃▂▅▇█▃▃▁ beats 36/60	hard · avg 28%	▇▇▇ 7.3%	at its level · small run

web search█··········█ #1 of 2 · 17% of its score · 3 tests

test	result	vs the others in its run	difficulty	counts	why
BrowseComp lab	67.6% ✎	▂·▄▁▂▃▆▄█▂▃▄ beats 30/42	hard · avg 23%	▇▇▇ 7.0%	at its level · small run
Humanity's Last Exam with tools lab	33.4% ✎	▂▂▂▁▄▂▃▆█▇▂▆ beats 12/54	very hard · avg 13%	▇▇ 5.4%	at its level · small run
WideSearch lab	67.8% ✎	█·▂▆··▂···▂▂ beats 6/10	medium · avg 35%	▇▇ 4.5%	at its level · small run

other tests3 not in a task · 31% of its score

test	result	vs the others in its run	difficulty	counts	why
GPQA Diamond lab	89.2% ✎	▁▁▁▁▂▄▅▅███▄ beats 258/310	hard · avg 45%	▇▇▇▇▇▇ 15%	sharp · small run
Humanity's Last Exam lab	25.6% ✎	▁▁▁▃▆▆▇██▆▁▃ beats 61/104	very hard · avg 4.8%	▇▇▇▇ 8.8%	at its level · small run
NL2Repo lab	46.2% ✎	▂▂·▂▂▄█▆▆▂▄▆ beats 15/21	very hard · avg 11%	▇▇▇ 7.0%	at its level · small run

$ where to run it

llama.cpp · recommended

version

Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
Windows
irm https://llama.app/install.ps1 | iex
other ways to install ↗
Run Ornith 1.5 35B-A3B · needs ~21.7 GB · from huggingface.co/ornith-ai (official)
chat in the terminal
llama cli -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Ornith 1.5 35B-A3B · needs ~21.7 GB · from huggingface.co/ornith-ai (official)
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Ollama

Install Ollama
download it (macOS / Windows) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Ornith 1.5 35B-A3B · needs ~23 GB · from ollama.com
chat in the terminal
ollama run ornith-1.5:35b

MLX (Mac)

version

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Ornith 1.5 35B-A3B · needs ~19.5 GB, Mac only · from huggingface.co/ornith-ai (official)
chat in the terminal
mlx_lm.chat --model ornith-ai/Ornith-1.5-35B-A3B-MLX-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model ornith-ai/Ornith-1.5-35B-A3B-MLX-4bit

Official weights and documentation: huggingface.co/ornith-ai ↗

$ how the score was worked out · ornith-ai/Ornith-1.5-35B-A3B… 99%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

source	results	share of the score	licence
ornith-ai/Ornith-1.5-35B-A3B model card · lab-reported	12	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 99%	MIT
prior · pull toward the middle		▇ 1%
weighted average		= 70.6

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated 2026-10-11 02:20 UTC · sources and licences · json