Update README.md
Browse files
README.md
CHANGED
|
@@ -16,11 +16,10 @@ tags:
|
|
| 16 |
|
| 17 |
# FastApply-1.5B-v1.0
|
| 18 |
*🚀 Update May 2025:* For production-grade throughput, we use *[Morph](https://morphllm.com)* (the hosted Fast Apply API powering [SoftGen AI](https://softgen.ai/)).
|
| 19 |
-
- Morph hits *~
|
| 20 |
- Larger model trained on millions of examples and tuned for accuracy.
|
| 21 |
> Stable inference, large free tier, highly recommended if you need serious speed in prod.
|
| 22 |
|
| 23 |
-
|
| 24 |
[Github: kortix-ai/fast-apply](https://github.com/kortix-ai/fast-apply)
|
| 25 |
[Dataset: Kortix/FastApply-dataset-v1.0](https://huggingface.co/datasets/Kortix/FastApply-dataset-v1.0)
|
| 26 |
[Try it now on 👉 Google Colab](https://colab.research.google.com/drive/1BNCab4oK-xBqwFQD4kCcjKc7BPKivkm1?usp=sharing)
|
|
|
|
| 16 |
|
| 17 |
# FastApply-1.5B-v1.0
|
| 18 |
*🚀 Update May 2025:* For production-grade throughput, we use *[Morph](https://morphllm.com)* (the hosted Fast Apply API powering [SoftGen AI](https://softgen.ai/)).
|
| 19 |
+
- Morph hits *~10,500 tok/s* even on huge token diffs 44k input
|
| 20 |
- Larger model trained on millions of examples and tuned for accuracy.
|
| 21 |
> Stable inference, large free tier, highly recommended if you need serious speed in prod.
|
| 22 |
|
|
|
|
| 23 |
[Github: kortix-ai/fast-apply](https://github.com/kortix-ai/fast-apply)
|
| 24 |
[Dataset: Kortix/FastApply-dataset-v1.0](https://huggingface.co/datasets/Kortix/FastApply-dataset-v1.0)
|
| 25 |
[Try it now on 👉 Google Colab](https://colab.research.google.com/drive/1BNCab4oK-xBqwFQD4kCcjKc7BPKivkm1?usp=sharing)
|