KAT-Coder-V2 Technical Report
Paper • 2603.27703 • Published • 9
AI & ML interests
Breaking the opacity of language models for legal professionals 📖 Join us by smashing the button at top right 🤗
Recent Activity
View all activity
AdinaY
submitted a
paper to Daily Papers 12 days ago
1aurent
authored a
paper 13 days ago
umarbutler
posted an update 22 days ago
Post
4834
Isaacus, the AI research company building legal superintelligence, is hiring!
We're looking for passionate engineers who love to build and tinker and want to have an impact on the world. Specifically, we're hiring:
• ML engineers (Australia).
• Data engineers (Australia).
• Full-stack engineers (Australia).
• DevRel engineers (Australia, San Francisco, and London).
• DevOps engineers (Australia, San Francisco, and London).
If you'd like to be a founding employee at one of the few VC-backed LLM research labs in the world, receive generous equity compensation, and work alongside other highly motivated, highly skilled engineers, get in touch: https://isaacus.com/careers
We're looking for passionate engineers who love to build and tinker and want to have an impact on the world. Specifically, we're hiring:
• ML engineers (Australia).
• Data engineers (Australia).
• Full-stack engineers (Australia).
• DevRel engineers (Australia, San Francisco, and London).
• DevOps engineers (Australia, San Francisco, and London).
If you'd like to be a founding employee at one of the few VC-backed LLM research labs in the world, receive generous equity compensation, and work alongside other highly motivated, highly skilled engineers, get in touch: https://isaacus.com/careers
Post
6521
We should really have a release date range slider on the /models page. Tired of "trending/most downloaded" being the best way to sort and still seeing models from 2023 on the first page just because they're embedded in enterprise pipelines and get downloaded repeatedly. "Recently Created/Recently Updated" don't solve the discovery problem considering the amount of noise to sift through.
Slight caveat: Trending actually does have some recency bias, but it's not strong/precise enough.
Slight caveat: Trending actually does have some recency bias, but it's not strong/precise enough.
- 3 replies
AdinaY
submitted 2
papers to Daily Papers about 1 month ago
umarbutler
posted an update about 1 month ago
Post
1944
This awesome visualization by @abdurrahmanbutler tracks how reliant the High Court of Australia has been on UK precedents over time.
Back in the early 1900s, up to 70% of citations in High Court decisions were from the UK. Today, that number sits around 20%.
This change seems to have happened gradually as Australia gained more and more independence from the UK, culminating in the Australia Acts of 1986, where we see a nice bump in the proportion of Australian cases cited.
These insights would not be possible without our latest legal AI model, Kanon 2 Enricher, which we used to extract dates and citations from High Court decisions in isaacus/open-australian-legal-corpus and categorize citations by jurisdiction. You can learn about Kanon 2 Enricher here: https://isaacus.com/blog/kanon-2-enricher.
Back in the early 1900s, up to 70% of citations in High Court decisions were from the UK. Today, that number sits around 20%.
This change seems to have happened gradually as Australia gained more and more independence from the UK, culminating in the Australia Acts of 1986, where we see a nice bump in the proportion of Australian cases cited.
These insights would not be possible without our latest legal AI model, Kanon 2 Enricher, which we used to extract dates and citations from High Court decisions in isaacus/open-australian-legal-corpus and categorize citations by jurisdiction. You can learn about Kanon 2 Enricher here: https://isaacus.com/blog/kanon-2-enricher.
umarbutler
authored a
paper about 1 month ago
umarbutler
submitted a
paper to Daily Papers about 1 month ago
Post
3537
🤔 Who would win ?
- a fully subsidized ai lab
- 3 random students named
kurakurai ?
demo : Tonic/fr-on-device
if you like it give the demo a little star and send a shoutout to : @MaxLSB @jddqd and @GAD-cell for absolutely obliterating the pareto frontier of the french language understanding .
- a fully subsidized ai lab
OR - 3 random students named
demo : Tonic/fr-on-device
if you like it give the demo a little star and send a shoutout to : @MaxLSB @jddqd and @GAD-cell for absolutely obliterating the pareto frontier of the french language understanding .
- 4 replies
umarbutler
posted an update about 2 months ago
Post
2192
@abdurrahmanbutler and I just dropped Legal RAG Bench, the first benchmark for legal RAG systems to simultaneously evaluate hallucinations, retrieval failures, and reasoning errors.
Our key takeaways are:
1. Embedding models, not generative models, are the primary driver of RAG accuracy. Switching from a general-purpose embedder like OpenAI's Text Embedding 3 Large to a legal domain embedder like Isaacus' Kanon 2 Embedder can raise accuracy by ~19 points.
2. Hallucinations are often triggered by retrieval failures. Fix your retrieval stack, and, in most cases, you end up fixing hallucinations.
3. Once you have a solid legal retrieval engine like Kanon 2 Embedder, it doesn’t matter as much what generative model you use; GPT-5.2 and Gemini 3.1 Pro perform relatively similarly, with Gemini 3.1 Pro achieving slightly better accuracy at the cost of more hallucinations.
4. Google's latest LLM, Gemini 3.1 Pro, is actually a bit worse than its predecessor at legal RAG, achieving 79.3% accuracy instead of 80.3%.
These findings confirm what we already knew at Isaacus: that information retrieval sets the ceiling on the accuracy of legal RAG systems. It doesn’t matter how smart you are; you aren’t going to magically know what the penalty is for speeding in California without access to an up-to-date copy of the California Vehicle Code.
Even still, to our knowledge, we’re the first to actually show this empirically.
Unfortunately, as we highlight in our write-up, high-quality open legal benchmarks like Legal RAG Bench and our earlier MLEB are few and far between.
In the interests of transparency, we have not only detailed exactly how we built Legal RAG Bench, but we’ve also released all of our data openly on Hugging Face. You can read our write up [here](https://isaacus.com/blog/legal-rag-bench), noting that we’ll soon be publishing it as a paper.
Kudos to my brother @abdurrahmanbutler for serving as the lead author on this monumental release.
Our key takeaways are:
1. Embedding models, not generative models, are the primary driver of RAG accuracy. Switching from a general-purpose embedder like OpenAI's Text Embedding 3 Large to a legal domain embedder like Isaacus' Kanon 2 Embedder can raise accuracy by ~19 points.
2. Hallucinations are often triggered by retrieval failures. Fix your retrieval stack, and, in most cases, you end up fixing hallucinations.
3. Once you have a solid legal retrieval engine like Kanon 2 Embedder, it doesn’t matter as much what generative model you use; GPT-5.2 and Gemini 3.1 Pro perform relatively similarly, with Gemini 3.1 Pro achieving slightly better accuracy at the cost of more hallucinations.
4. Google's latest LLM, Gemini 3.1 Pro, is actually a bit worse than its predecessor at legal RAG, achieving 79.3% accuracy instead of 80.3%.
These findings confirm what we already knew at Isaacus: that information retrieval sets the ceiling on the accuracy of legal RAG systems. It doesn’t matter how smart you are; you aren’t going to magically know what the penalty is for speeding in California without access to an up-to-date copy of the California Vehicle Code.
Even still, to our knowledge, we’re the first to actually show this empirically.
Unfortunately, as we highlight in our write-up, high-quality open legal benchmarks like Legal RAG Bench and our earlier MLEB are few and far between.
In the interests of transparency, we have not only detailed exactly how we built Legal RAG Bench, but we’ve also released all of our data openly on Hugging Face. You can read our write up [here](https://isaacus.com/blog/legal-rag-bench), noting that we’ll soon be publishing it as a paper.
Kudos to my brother @abdurrahmanbutler for serving as the lead author on this monumental release.
- 2 replies
Post
3353
🙋🏻♂️hello my lovelies ,
it is with great pleasure i present to you my working one-click deploy 16GB ram completely free huggingface spaces deployment.
repo : Tonic/hugging-claw (use git clone to inspect)
literally the one-click link : Tonic/hugging-claw
you can also run it locally and see for yourself :
docker run -it -p 7860:7860 --platform=linux/amd64 \
-e HF_TOKEN="YOUR_VALUE_HERE" \
-e OPENCLAW_GATEWAY_TRUSTED_PROXIES="YOUR_VALUE_HERE" \
-e OPENCLAW_GATEWAY_PASSWORD="YOUR_VALUE_HERE" \
-e OPENCLAW_CONTROL_UI_ALLOWED_ORIGINS="YOUR_VALUE_HERE" \
registry.hf.space/tonic-hugging-claw:latest
just a few quite minor details i'll take care of but i wanted to share here first
it is with great pleasure i present to you my working one-click deploy 16GB ram completely free huggingface spaces deployment.
repo : Tonic/hugging-claw (use git clone to inspect)
literally the one-click link : Tonic/hugging-claw
you can also run it locally and see for yourself :
docker run -it -p 7860:7860 --platform=linux/amd64 \
-e HF_TOKEN="YOUR_VALUE_HERE" \
-e OPENCLAW_GATEWAY_TRUSTED_PROXIES="YOUR_VALUE_HERE" \
-e OPENCLAW_GATEWAY_PASSWORD="YOUR_VALUE_HERE" \
-e OPENCLAW_CONTROL_UI_ALLOWED_ORIGINS="YOUR_VALUE_HERE" \
registry.hf.space/tonic-hugging-claw:latest
just a few quite minor details i'll take care of but i wanted to share here first
- 2 replies
Post
3569
MiniMax M2.5 is now available on the hub 🚀
MiniMaxAI/MiniMax-M2.5
✨ 229B - Modified MIT license
✨37% faster than M2.1
✨ ~$1/hour at 100 TPS
MiniMaxAI/MiniMax-M2.5
✨ 229B - Modified MIT license
✨37% faster than M2.1
✨ ~$1/hour at 100 TPS
- 2 replies
Post
726
RynnBrain 🤖 a physics aware embodied brain for robots from Alibaba DAMO
https://huggingface.co/collections/Alibaba-DAMO-Academy/rynnbrain
✨ 2B/8B/30B (3B active)
✨ Apache 2.0
✨ Understands egocentric scenes with strong spatial awareness
✨ Tracks objects and motion over time
https://huggingface.co/collections/Alibaba-DAMO-Academy/rynnbrain
✨ 2B/8B/30B (3B active)
✨ Apache 2.0
✨ Understands egocentric scenes with strong spatial awareness
✨ Tracks objects and motion over time
- 2 replies
umarbutler
posted an update about 2 months ago
Post
5080
What happens when you annotate, extract, and disambiguate every entity mentioned in the longest U.S. Supreme Court decision in history? What if you then linked those entities to each other and visualized it as a network?
This is the result of enriching all 241 pages and 111,267 words of Dred Scott v. Sandford (1857) with Kanon 2 Enricher in less than ten seconds at the cost of 47 cents.
Dred Scott v. Sandford is the longest U.S. Supreme Court decision by far, and has variously been called "the worst Supreme Court decision ever" and "the Court's greatest self-inflicted wound" due to its denial of the rights of African Americans.
Thanks to Kanon 2 Enricher, we now also know that the case contains 950 numbered paragraphs, 6 footnotes, 178 people mentioned 1,340 times, 99 locations mentioned 1,294 times, and 298 external documents referenced 940 times.
For an American case, there are a decent number of references to British precedents (27 to be exact), including the Magna Carta (¶ 928).
Surprisingly though, the Magna Carta is not the oldest citation referenced. That would be the Institutes of Justinian (¶ 315), dated around 533 CE.
The oldest city mentioned is Rome (founded 753 BCE) (¶ 311), the oldest person is Justinian (born 527 CE) (¶ 314), and the oldest year referenced is 1371, when 'Charles V of France exempted all the inhabitants of Paris from serfdom' (¶ 370).
All this information and more was extracted in 9 seconds. That's how powerful Kanon 2 Enricher, my latest LLM for document enrichment and hierarchical graphitization, is. If you'd like to play with it yourself now that it's available in closed beta, you can apply to the Isaacus Beta Program here: https://isaacus.com/beta.
This is the result of enriching all 241 pages and 111,267 words of Dred Scott v. Sandford (1857) with Kanon 2 Enricher in less than ten seconds at the cost of 47 cents.
Dred Scott v. Sandford is the longest U.S. Supreme Court decision by far, and has variously been called "the worst Supreme Court decision ever" and "the Court's greatest self-inflicted wound" due to its denial of the rights of African Americans.
Thanks to Kanon 2 Enricher, we now also know that the case contains 950 numbered paragraphs, 6 footnotes, 178 people mentioned 1,340 times, 99 locations mentioned 1,294 times, and 298 external documents referenced 940 times.
For an American case, there are a decent number of references to British precedents (27 to be exact), including the Magna Carta (¶ 928).
Surprisingly though, the Magna Carta is not the oldest citation referenced. That would be the Institutes of Justinian (¶ 315), dated around 533 CE.
The oldest city mentioned is Rome (founded 753 BCE) (¶ 311), the oldest person is Justinian (born 527 CE) (¶ 314), and the oldest year referenced is 1371, when 'Charles V of France exempted all the inhabitants of Paris from serfdom' (¶ 370).
All this information and more was extracted in 9 seconds. That's how powerful Kanon 2 Enricher, my latest LLM for document enrichment and hierarchical graphitization, is. If you'd like to play with it yourself now that it's available in closed beta, you can apply to the Isaacus Beta Program here: https://isaacus.com/beta.
Post
4084
Game on 🎮🚀
While Seedance 2.0’s videos are all over the timeline, DeepSeek quietly pushed a new model update in its app.
GLM-5 from Z.ai adds more momentum.
Ming-flash-omni from Ant Group , MiniCPM-SALA from OpenBMB
, and the upcoming MiniMax M2.5 keep the heat on 🔥
Spring Festival is around the corner,
no one’s sleeping!
✨ More releases coming, stay tuned
https://huggingface.co/collections/zh-ai-community/2026-february-china-open-source-highlights
While Seedance 2.0’s videos are all over the timeline, DeepSeek quietly pushed a new model update in its app.
GLM-5 from Z.ai adds more momentum.
Ming-flash-omni from Ant Group , MiniCPM-SALA from OpenBMB
, and the upcoming MiniMax M2.5 keep the heat on 🔥
Spring Festival is around the corner,
no one’s sleeping!
✨ More releases coming, stay tuned
https://huggingface.co/collections/zh-ai-community/2026-february-china-open-source-highlights
Post
3940
Ming-flash-omni 2.0 🚀 New open omni-MLLM released by Ant Group
inclusionAI/Ming-flash-omni-2.0
✨ MIT license
✨ MoE - 100B/6B active
✨ Zero-shot voice cloning + controllable audio
✨ Fine-grained visual knowledge grounding
inclusionAI/Ming-flash-omni-2.0
✨ MIT license
✨ MoE - 100B/6B active
✨ Zero-shot voice cloning + controllable audio
✨ Fine-grained visual knowledge grounding
- 2 replies
1aurent
authored a
paper 2 months ago
alokabhishek
submitted a
paper to Daily Papers 2 months ago
Post
796
LLaDA 2.1 is out 🔥 A new series of MoE diffusion language model released by AntGroup
inclusionAI/LLaDA2.1-mini
inclusionAI/LLaDA2.1-flash
✨LLaDA2.1-mini: 16B - Apache2.0
✨LLaDA2.1-flash: 100B - Apache2.0
✨Both delivers editable generation, RL-trained diffusion reasoning and fast inference
inclusionAI/LLaDA2.1-mini
inclusionAI/LLaDA2.1-flash
✨LLaDA2.1-mini: 16B - Apache2.0
✨LLaDA2.1-flash: 100B - Apache2.0
✨Both delivers editable generation, RL-trained diffusion reasoning and fast inference
- 2 replies