AxiomicLabs/GPT-X2.5-135M
Text Generation • 0.1B • Updated • 1.63k • 60
Dean Byrne PRO
Quazim0t0
RebelBroUncensored's profile picture
juiceb0xc0de's profile picture
SoulInPsyAbstract's profile picture·
AI & ML interests
DaisyChainAI🌼 / SmallLM's / San Francisco / Open Source
Recent Activity
liked a model 7 days ago
AxiomicLabs/GPT-X2.5-135M reacted to AbstractPhil's post with 🔥 8 days ago
The AlephLM results are rolling in and I'm very excited for the possibilities. I am very much looking forward to the coming weeks as I train the first AlephLM distillations from MANY teachers into AMOE arms.
The AMOE arms hook cleanly to AlephLM structures and provide pos/neg learning elements. Hard positive and hard negatives coalesce to extend the capacity.
https://huggingface.co/AbstractPhil/alephlm-0
https://huggingface.co/AbstractPhil/alephlm-adopt-0
As it stands they are structurally sound enough to fully pretrain. As or more stable than a standard Bert experimentally to distill using InfoNCE. AMOE legs improve these structures substantially.
Structural behavior can be expanded in many ways on distilled and pretrained models alike. Attaching the AMOE to any model I've tried has created expanded or improved behavioral accumulations. They do have downsides but their upsides are very experimentally exciting.
I've distilled multiple vits, multiple berts, and have begun distilling berts into AlephLM structures successfully.
This is overall very exciting for me. I've begun formatting larger variants such as including GPT-2 and Qwen 3.5 4b as a paired combinator utilizing pathological T5 learned distilled encodings. It sounds odd, but the results show everything can be expanded and even be taught to cooperate.
The CaptionBert-8192-v2 and v2-b are both structurally collapsing after token 480 or so, which is expected due to the small train. By distilling an AMOE arm to V2 by training with a longformer expert, the results are cutting through like butter. V2 has begun stabilizing rapidly for considerably longer token chains and sequences, the structure is repairing and building reusable capacity.
I have discovered an improved methodology for sampling the AlephLM for text encoder benchmarks, which is predominantly L2 normalized outputs.
Upcoming large paper for the distillation experiments and results within the next week or two. It's going to be a big one. updated a model 8 days ago
Quazim0t0/CS2DreamOrganizations
liked a
model 7 days ago
reacted to AbstractPhil's post with 🔥 8 days ago
Post
2721
The AlephLM results are rolling in and I'm very excited for the possibilities. I am very much looking forward to the coming weeks as I train the first AlephLM distillations from MANY teachers into AMOE arms.
The AMOE arms hook cleanly to AlephLM structures and provide pos/neg learning elements. Hard positive and hard negatives coalesce to extend the capacity.
AbstractPhil/alephlm-0
AbstractPhil/alephlm-adopt-0
As it stands they are structurally sound enough to fully pretrain. As or more stable than a standard Bert experimentally to distill using InfoNCE. AMOE legs improve these structures substantially.
Structural behavior can be expanded in many ways on distilled and pretrained models alike. Attaching the AMOE to any model I've tried has created expanded or improved behavioral accumulations. They do have downsides but their upsides are very experimentally exciting.
I've distilled multiple vits, multiple berts, and have begun distilling berts into AlephLM structures successfully.
This is overall very exciting for me. I've begun formatting larger variants such as including GPT-2 and Qwen 3.5 4b as a paired combinator utilizing pathological T5 learned distilled encodings. It sounds odd, but the results show everything can be expanded and even be taught to cooperate.
The CaptionBert-8192-v2 and v2-b are both structurally collapsing after token 480 or so, which is expected due to the small train. By distilling an AMOE arm to V2 by training with a longformer expert, the results are cutting through like butter. V2 has begun stabilizing rapidly for considerably longer token chains and sequences, the structure is repairing and building reusable capacity.
I have discovered an improved methodology for sampling the AlephLM for text encoder benchmarks, which is predominantly L2 normalized outputs.
Upcoming large paper for the distillation experiments and results within the next week or two. It's going to be a big one.
The AMOE arms hook cleanly to AlephLM structures and provide pos/neg learning elements. Hard positive and hard negatives coalesce to extend the capacity.
AbstractPhil/alephlm-0
AbstractPhil/alephlm-adopt-0
As it stands they are structurally sound enough to fully pretrain. As or more stable than a standard Bert experimentally to distill using InfoNCE. AMOE legs improve these structures substantially.
Structural behavior can be expanded in many ways on distilled and pretrained models alike. Attaching the AMOE to any model I've tried has created expanded or improved behavioral accumulations. They do have downsides but their upsides are very experimentally exciting.
I've distilled multiple vits, multiple berts, and have begun distilling berts into AlephLM structures successfully.
This is overall very exciting for me. I've begun formatting larger variants such as including GPT-2 and Qwen 3.5 4b as a paired combinator utilizing pathological T5 learned distilled encodings. It sounds odd, but the results show everything can be expanded and even be taught to cooperate.
The CaptionBert-8192-v2 and v2-b are both structurally collapsing after token 480 or so, which is expected due to the small train. By distilling an AMOE arm to V2 by training with a longformer expert, the results are cutting through like butter. V2 has begun stabilizing rapidly for considerably longer token chains and sequences, the structure is repairing and building reusable capacity.
I have discovered an improved methodology for sampling the AlephLM for text encoder benchmarks, which is predominantly L2 normalized outputs.
Upcoming large paper for the distillation experiments and results within the next week or two. It's going to be a big one.
- 3 replies
updated a
model 8 days ago
reacted to salma-remyx's post with 🤗 8 days ago
Post
2783
I unblocked 15 improvements for VQASynth in a day!
VQASynth, the pipeline behind SpaceThinker-Qwen2.5VL-3B, SpaceLLaVA, and the SpaceThinker dataset, had a backlog spanning new annotation methods, curation pipelines, model integrations, and agent tools.
I scoped each item in a short design brief. In GitHub Actions, Outrider mapped them to the repo’s existing modules and interfaces, implemented the changes, added tests, ran the repo checks, and opened draft PRs.
The 15 PRs covered:
* object orientation and 3D bounding boxes
* SAM2 regional captioning and multi-view matching
* LLaMA-Mesh tokenization
* Qwen2.5-VL fine-tuning
* spatial-reasoning data generation
* CLIP, SigLIP, and LLM2CLIP backends
and more
The batch added 12,275 lines, nearly doubling the codebase.
SpatialAnnotator went from two tools to seven, creating combinatorially more possible annotation pipelines for each image.
Outrider handled the implementation, tests, and integration. At about $1 per branch, I spent my time reviewing 15 concrete changes, fixing what needed fixing, and deciding what to ship.
Outrider is open source!
Point it at a paper, issue, or design brief and review the PR: https://github.com/remyxai/outrider
Check out the updates on VQASynth: https://github.com/remyxai/VQASynth
VQASynth, the pipeline behind SpaceThinker-Qwen2.5VL-3B, SpaceLLaVA, and the SpaceThinker dataset, had a backlog spanning new annotation methods, curation pipelines, model integrations, and agent tools.
I scoped each item in a short design brief. In GitHub Actions, Outrider mapped them to the repo’s existing modules and interfaces, implemented the changes, added tests, ran the repo checks, and opened draft PRs.
The 15 PRs covered:
* object orientation and 3D bounding boxes
* SAM2 regional captioning and multi-view matching
* LLaMA-Mesh tokenization
* Qwen2.5-VL fine-tuning
* spatial-reasoning data generation
* CLIP, SigLIP, and LLM2CLIP backends
and more
The batch added 12,275 lines, nearly doubling the codebase.
SpatialAnnotator went from two tools to seven, creating combinatorially more possible annotation pipelines for each image.
Outrider handled the implementation, tests, and integration. At about $1 per branch, I spent my time reviewing 15 concrete changes, fixing what needed fixing, and deciding what to ship.
Outrider is open source!
Point it at a paper, issue, or design brief and review the PR: https://github.com/remyxai/outrider
Check out the updates on VQASynth: https://github.com/remyxai/VQASynth
reacted to Banaxi-Tech's post with ❤️🤗 8 days ago
Post
2317
We're excited to release BananaMind 2 Pro Preview, our best model yet.
Trained on ~52B tokens it performs extremely good for its token and size class.
We trained it on a single 5070 Ti in about 11 days.
Check it out at BananaMind/BananaMind-2-Pro-Preview.
Sadly we need to delay BananaMind 2 Micro until the launch of the final BananaMind 2 Pro.
We will release the final checkpoint with 100B tokens in ~11 days.
Go and fine-tune it!
We've also released BananaMind 2 Pro Preview Chat which is the instruct version of it!
BananaMind/BananaMind-2-Pro-Preview-Chat
Follow us to know when the final releases and support us at
BananaMind
@Banaxi-Tech
Trained on ~52B tokens it performs extremely good for its token and size class.
We trained it on a single 5070 Ti in about 11 days.
Check it out at BananaMind/BananaMind-2-Pro-Preview.
Sadly we need to delay BananaMind 2 Micro until the launch of the final BananaMind 2 Pro.
We will release the final checkpoint with 100B tokens in ~11 days.
Go and fine-tune it!
We've also released BananaMind 2 Pro Preview Chat which is the instruct version of it!
BananaMind/BananaMind-2-Pro-Preview-Chat
Follow us to know when the final releases and support us at
@Banaxi-Tech
- 5 replies
reacted to salma-remyx's post with 🚀 8 days ago
Post
2783
I unblocked 15 improvements for VQASynth in a day!
VQASynth, the pipeline behind SpaceThinker-Qwen2.5VL-3B, SpaceLLaVA, and the SpaceThinker dataset, had a backlog spanning new annotation methods, curation pipelines, model integrations, and agent tools.
I scoped each item in a short design brief. In GitHub Actions, Outrider mapped them to the repo’s existing modules and interfaces, implemented the changes, added tests, ran the repo checks, and opened draft PRs.
The 15 PRs covered:
* object orientation and 3D bounding boxes
* SAM2 regional captioning and multi-view matching
* LLaMA-Mesh tokenization
* Qwen2.5-VL fine-tuning
* spatial-reasoning data generation
* CLIP, SigLIP, and LLM2CLIP backends
and more
The batch added 12,275 lines, nearly doubling the codebase.
SpatialAnnotator went from two tools to seven, creating combinatorially more possible annotation pipelines for each image.
Outrider handled the implementation, tests, and integration. At about $1 per branch, I spent my time reviewing 15 concrete changes, fixing what needed fixing, and deciding what to ship.
Outrider is open source!
Point it at a paper, issue, or design brief and review the PR: https://github.com/remyxai/outrider
Check out the updates on VQASynth: https://github.com/remyxai/VQASynth
VQASynth, the pipeline behind SpaceThinker-Qwen2.5VL-3B, SpaceLLaVA, and the SpaceThinker dataset, had a backlog spanning new annotation methods, curation pipelines, model integrations, and agent tools.
I scoped each item in a short design brief. In GitHub Actions, Outrider mapped them to the repo’s existing modules and interfaces, implemented the changes, added tests, ran the repo checks, and opened draft PRs.
The 15 PRs covered:
* object orientation and 3D bounding boxes
* SAM2 regional captioning and multi-view matching
* LLaMA-Mesh tokenization
* Qwen2.5-VL fine-tuning
* spatial-reasoning data generation
* CLIP, SigLIP, and LLM2CLIP backends
and more
The batch added 12,275 lines, nearly doubling the codebase.
SpatialAnnotator went from two tools to seven, creating combinatorially more possible annotation pipelines for each image.
Outrider handled the implementation, tests, and integration. At about $1 per branch, I spent my time reviewing 15 concrete changes, fixing what needed fixing, and deciding what to ship.
Outrider is open source!
Point it at a paper, issue, or design brief and review the PR: https://github.com/remyxai/outrider
Check out the updates on VQASynth: https://github.com/remyxai/VQASynth
reacted to mlabonne's post with 🚀 8 days ago
Post
3137
Deploy local agents everywhere with LiquidAI/LFM2.5-2.6B
Unlike agents that depend on cloud APIs, local agents give you free inference, low latency, and real privacy.
Removing the per-token cost changes how developers build: agents can now be massively parallelized on local hardware, running background tasks that burn through millions of tokens at no marginal cost!
Unlike agents that depend on cloud APIs, local agents give you free inference, low latency, and real privacy.
Removing the per-token cost changes how developers build: agents can now be massively parallelized on local hardware, running background tasks that burn through millions of tokens at no marginal cost!
- 7 replies
reacted to Enderchef's post with ❤️🔥🤝 8 days ago
Post
3408
🚀 Supra2 100M is out, and multiple other SLM orgs are gaining power!
Following takes a press. Please follow:
fromziro
SupraLabs
AxiomicLabs
Following takes a press. Please follow:
- 8 replies
reacted to Enderchef's post with 🤗 8 days ago
Post
3247
GPT-X2.5-135M is finally released! 🚀
The new flagship from Axiomic Labs takes 3rd on the open SLM leaderboard trailing only the SmolLMs, check it out and follow us:
AxiomicLabs/GPT-X2.5-135M
The new flagship from Axiomic Labs takes 3rd on the open SLM leaderboard trailing only the SmolLMs, check it out and follow us:
AxiomicLabs/GPT-X2.5-135M
- 2 replies
reacted to HannesVonEssen's post with 🚀🔥🤗❤️ 8 days ago
Post
2317
🏅 HF Viewer podium
Our top most active users:
🥇#1 @Quazim0t0
🥈#2 @IvmeLabs
🥉#3 @GODELEV
If you are an HF creator and want to try entering the podium, feel free to visit the embed page to add your graph!
https://hfviewer.com/model-card-embed
Our top most active users:
🥇#1 @Quazim0t0
🥈#2 @IvmeLabs
🥉#3 @GODELEV
If you are an HF creator and want to try entering the podium, feel free to visit the embed page to add your graph!
https://hfviewer.com/model-card-embed
- 2 replies
replied to HannesVonEssen's post 8 days ago
Awesome! I'm glad to be using it! I think it will become the standard for when people do a showcase or project release.
reacted to Enderchef's post with 🔥 10 days ago
Post
164
New repo, new article!
Check out and give a star to(and spread the word, help me go viral) https://github.com/naxium-oss/DistribAI , and check out https://huggingface.co/blog/Enderchef/distribai-release for the full article and release documentation!
Check out and give a star to(and spread the word, help me go viral) https://github.com/naxium-oss/DistribAI , and check out https://huggingface.co/blog/Enderchef/distribai-release for the full article and release documentation!
liked a
dataset 11 days ago