Once you have tasted flight, you will walk the earth with your eyes turned skywards.
SF Bay Area
- Tweets 480
- Following 312
- Followers 18,976
- Likes 2,410
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies.
It achieves state-of-the-art performance among open-source models of comparable size and delivers performance comparable to Claude Opus 4.8 across reasoning, agentic, and coding tasks:
✅Terminal-Bench 2.1 (86.1)
✅SWE-Bench (86 on verified, 65.1 on pro, 79.6 on Multilingual)
✅DeepSWE (56)
✅HLE (44.6)
✅ClawEval (81.4)
✅Tool Decathlon (71.2)
Ornith-1.5 takes a major step toward training foundation models through end-to-end self-improvement, extending the self-scaffolding strategies introduced in Ornith-1.0 into a more complete self-improvement loop: the model proposes new tasks, generates task-specific scaffolds, and produces solution rollouts for reinforcement learning, continuously creating new learning experiences from which it can improve.
All models, along with their quantized versions (FP8, GGUF, MLX, and NVFP4), have been released under the MIT License, enabling unrestricted commercial and research use.
📘Tech Blog: ornith.ai/ornith_1_5.html
🤗Huggingface: huggingface.co/collections/o…
413
940
6,806
4,546,552
🔍We dove a bit deeper and analyzed how multi-token prediction (MTP) works in Ornith 1.5 models.
🪽With MTP, Ornith 1.5 gain inference speedups without quality loss via self-speculative decoding: models use their own MTP head to draft tokens and verify them in a forward pass.
🐦MTP weights are now updated for all 9B, 35B, and 397B models in BF16, GGUF, FP8, and NVFP4 variants (except 397B NVFP4, will be updated today).
huggingface.co/collections/o…
🐦Tokens now go brrrr
🪽Ornith-1.5 models just got new wings: MTP weights have been updated for the BF16, GGUF, NVFP4 & FP8 variants.
🔗huggingface.co/collections/o…
29
25
358
28,511
🐦It's use case time!
🧐Here's a nest of spicy use cases you might also find interesting.
🧵1/
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies.
It achieves state-of-the-art performance among open-source models of comparable size and delivers performance comparable to Claude Opus 4.8 across reasoning, agentic, and coding tasks:
✅Terminal-Bench 2.1 (86.1)
✅SWE-Bench (86 on verified, 65.1 on pro, 79.6 on Multilingual)
✅DeepSWE (56)
✅HLE (44.6)
✅ClawEval (81.4)
✅Tool Decathlon (71.2)
Ornith-1.5 takes a major step toward training foundation models through end-to-end self-improvement, extending the self-scaffolding strategies introduced in Ornith-1.0 into a more complete self-improvement loop: the model proposes new tasks, generates task-specific scaffolds, and produces solution rollouts for reinforcement learning, continuously creating new learning experiences from which it can improve.
All models, along with their quantized versions (FP8, GGUF, MLX, and NVFP4), have been released under the MIT License, enabling unrestricted commercial and research use.
📘Tech Blog: ornith.ai/ornith_1_5.html
🤗Huggingface: huggingface.co/collections/o…
7
10
101
10,135
🧐To help everyone easily adopt advanced intelligence and empower everyday tasks, we also released FP8, NVFP4, GGUF, and MLX variants of the Ornith 1.5 series models.
🐦Blog: ornith.ai/ornith_1_5.html
🐦HF: huggingface.co/collections/o…
2
15
841
🐦Tokens now go brrrr
🪽Ornith-1.5 models just got new wings: MTP weights have been updated for the BF16, GGUF, NVFP4 & FP8 variants.
🔗huggingface.co/collections/o…
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies.
It achieves state-of-the-art performance among open-source models of comparable size and delivers performance comparable to Claude Opus 4.8 across reasoning, agentic, and coding tasks:
✅Terminal-Bench 2.1 (86.1)
✅SWE-Bench (86 on verified, 65.1 on pro, 79.6 on Multilingual)
✅DeepSWE (56)
✅HLE (44.6)
✅ClawEval (81.4)
✅Tool Decathlon (71.2)
Ornith-1.5 takes a major step toward training foundation models through end-to-end self-improvement, extending the self-scaffolding strategies introduced in Ornith-1.0 into a more complete self-improvement loop: the model proposes new tasks, generates task-specific scaffolds, and produces solution rollouts for reinforcement learning, continuously creating new learning experiences from which it can improve.
All models, along with their quantized versions (FP8, GGUF, MLX, and NVFP4), have been released under the MIT License, enabling unrestricted commercial and research use.
📘Tech Blog: ornith.ai/ornith_1_5.html
🤗Huggingface: huggingface.co/collections/o…
58
62
862
92,520
🫡
turned thinking mode on and re-ran the hard test on Ornith 1.5 35B.
GPQA-diamond, same harness as this morning's board, zero-shot greedy, 16k reasoning budget: 52.0 -> 81.8.
that lands it level with the best dense 27B rungs on my ladder with their thinking on (79.3 to 80.8), and this model decodes at 303 tok/s on one RTX 5090 where those rungs do 53 to 96.
thinking mode is worth ~30 points on this benchmark, the same band the dense ladder measured.
7
5
131
9,886
🐦Now we can just burn some tokens🔥
🫡Thanks @runinfrai for the support!
runinfra.ai/inference-api/or…
216 tok/s. 167ms to first token. 262,144 token context
Ornith 1.5 35B is now live on RunInfra
$0.10/1M input. $0.40/1M output. $0.01/1M cached. 90% of input hits the cache, so effective input is ~$0.02/1M
1M in + 1M out with a warm cache costs $0.42
runinfra.ai/inference-api/or…
8
5
156
12,058
🧡we love feedbacks!
Over the last day I've run an extensive set of benchmarks on Qwen3.6-35B-A3B vs Ornith-1.5-35B-A3B.
The results are in. I will post a more detailed thread of the breakdown shortly, but the clear winner is Ornith. Especially as it comes to coding, math and science.
I chose to test both MoE models against each other to compare apples to apples.
12
5
132
10,629
🐦it's time to download and play with Ornith🤗
huggingface.co/ornith-ai/Orn…
Ever wanted a compact AI that talks like a pro? Meet Ornith-1.5-9B-GGUF, a text-generation powerhouse that fits right into your workflow. With 54k downloads and 101 likes, it's already winning hearts. Curious? Let's dive in! #AI #LLM
16
8
118
9,360
🐦Now you can serve Ornith 1.5 with SGLang!
🫡Thanks for the support @sgl_project
Deployment commands 👇
Congrats to the @ornith_ team on launching the Ornith-1.5 family!
Very cool work on models that improve themselves, and thanks for including SGLang in the model cards🧡
Cookbook is on the way. Excited to see what people build and run with them!
Commands👇
13
8
98
10,673
👉Ornith-1.5-35B-A3B
Ornith-1.5-35B-A3B
python -m sglang.launch_server \
--model-path ornith-ai/Ornith-1.5-35B-A3B \
--served-model-name Ornith-1.5-35B-A3B \
--host 0.0.0.0 --port 8000 \
--tp 2 \
--context-length 262144 \
--mem-fraction-static 0.85 \
--tool-call-parser qwen3_coder \
--reasoning-parser qwen3
2
302
👉Ornith-1.5-397B:
Ornith-1.5-397B:
python -m sglang.launch_server \
--model-path ornith-ai/Ornith-1.5-397B \
--served-model-name Ornith-1.5-397B \
--host 0.0.0.0 --port 8000 \
--tp 8 \
--context-length 262144 \
--mem-fraction-static 0.85 \
--tool-call-parser qwen3_coder \
--reasoning-parser qwen3
7
2,120
🐦Ornith 1.5 series models are now available on ollama!
🫡Thanks to our awesome friends at @ollama
👀Come give Ornith 1.5 a try!
ollama.com/library/ornith-1.…
Aloha! 🌺Introducing Ornith-1.5, a family of open-source LLMs spanning 9B Dense, 35B MoE, and 397B MoE, trained with self-improving strategies.
It achieves state-of-the-art performance among open-source models of comparable size and delivers performance comparable to Claude Opus 4.8 across reasoning, agentic, and coding tasks:
✅Terminal-Bench 2.1 (86.1)
✅SWE-Bench (86 on verified, 65.1 on pro, 79.6 on Multilingual)
✅DeepSWE (56)
✅HLE (44.6)
✅ClawEval (81.4)
✅Tool Decathlon (71.2)
Ornith-1.5 takes a major step toward training foundation models through end-to-end self-improvement, extending the self-scaffolding strategies introduced in Ornith-1.0 into a more complete self-improvement loop: the model proposes new tasks, generates task-specific scaffolds, and produces solution rollouts for reinforcement learning, continuously creating new learning experiences from which it can improve.
All models, along with their quantized versions (FP8, GGUF, MLX, and NVFP4), have been released under the MIT License, enabling unrestricted commercial and research use.
📘Tech Blog: ornith.ai/ornith_1_5.html
🤗Huggingface: huggingface.co/collections/o…
13
25
187
18,148