TeamNull commited on
Commit
00f6fa0
Β·
verified Β·
1 Parent(s): c9689aa

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -5
README.md CHANGED
@@ -38,7 +38,8 @@ Our pretrained model series, trained from the ground up rather than fine-tuned.
38
  |---|---|---|---|---|
39
  | **Hydrion v1** | 114M | Base | Released | [Hugging Face β†’](https://huggingface.co/OPENGCM/Hydrion-v1-Base) |
40
  | **Hydrion v1-SFT** | 114M | Chat-ready | Released | [Hugging Face β†’](https://huggingface.co/OPENGCM/Hydrion-v1-SFT) |
41
- | **Hydrion v3.2** | ~250M | Chat-ready | In training |
 
42
 
43
 
44
  ---
@@ -76,7 +77,7 @@ Our current flagship GTM model is GTM-v2-chat.
76
  Small models don't have to mean small ambitions. We're building on a few principles:
77
 
78
  - **Open** β€” every weight, dataset, and training method we use is public, not just the final model
79
- - **Efficient** β€” Hydrion is trained on powerful GPUs, but can be ran on consumer GPUs
80
  - **Customizable** β€” models built to be forked, fine-tuned, and taken apart
81
  - **Accessible** β€” no gatekeeping behind closed weights or paywalled APIs
82
 
@@ -89,9 +90,6 @@ We're currently focused on:
89
  - **Pretraining from scratch** β€” Hydrion, our from-scratch language model series, and the data/architecture decisions that go into it
90
  - **Fine-tuning** β€” GCM, our code-focused chat models built on existing open-weight bases
91
  - **Evaluation** β€” benchmarking against public leaderboards and tracking what actually moves the needle
92
- - **Local deployment** β€” making sure what we build can run on real, non-datacenter hardware
93
-
94
- Follow along as it happens in our [Research space](https://huggingface.co/spaces/OPENGCM/Research).
95
 
96
  ---
97
 
 
38
  |---|---|---|---|---|
39
  | **Hydrion v1** | 114M | Base | Released | [Hugging Face β†’](https://huggingface.co/OPENGCM/Hydrion-v1-Base) |
40
  | **Hydrion v1-SFT** | 114M | Chat-ready | Released | [Hugging Face β†’](https://huggingface.co/OPENGCM/Hydrion-v1-SFT) |
41
+ | **Hydrion v2** | ~250M | Base | In training |
42
+ | **???????** | ~250M | Base | In training |
43
 
44
 
45
  ---
 
77
  Small models don't have to mean small ambitions. We're building on a few principles:
78
 
79
  - **Open** β€” every weight, dataset, and training method we use is public, not just the final model
80
+ - **Efficient** β€” our models are trained on powerful GPUs, but can be ran on consumer GPUs
81
  - **Customizable** β€” models built to be forked, fine-tuned, and taken apart
82
  - **Accessible** β€” no gatekeeping behind closed weights or paywalled APIs
83
 
 
90
  - **Pretraining from scratch** β€” Hydrion, our from-scratch language model series, and the data/architecture decisions that go into it
91
  - **Fine-tuning** β€” GCM, our code-focused chat models built on existing open-weight bases
92
  - **Evaluation** β€” benchmarking against public leaderboards and tracking what actually moves the needle
 
 
 
93
 
94
  ---
95