AICoder

AI Coding Live News

First-hand AI coding updates from 𝕏/Twitter frontier labs and researchers, distilled by AI with entity benchmarks.

🏆 Model leaderboard⚖️ Compare models💰 Compare plansExplore live on 𝕏 ↗

AI Coding Live News

P
Perplexity Developers@perplexitydevs·
👀19
🔥82
View on 𝕏 ↗

Perplexity Open-Sources pplx-decider v1.3 Decision Models: New Locate Span Extraction, 91.6% Mean Accuracy Across 63 Classification Tasks for the 26B Model, Apache 2.0, Free on OpenRouter Until November 19

On November 12 Perplexity released pplx-decider-v1.3 (built on Qwen3.8-27B, about 26.1B parameters) and the distilled pplx-decider-flash-v1.3 (built on Qwen3.5-4B, about 4.5B). The weights are on Hugging Face under Apache 2.0, and both models are served through the Decisions API and OpenRouter. The new Locate feature extracts verbatim spans by label or instruction and returns character offsets. Perplexity says v1.3 averages 91.6% accuracy across 63 classification tasks, 6.54 points above its production classifiers.

Perplexity Open-Sources pplx-decider v1.3 Decision Models: New Locate Span Extraction, 91.6% Mean Accuracy Across 63 Classification Tasks for the 26B Model, Apache 2.0, Free on OpenRouter Until November 19

Key Takeaways

  • Two open-weight sizes: v1.3 has about 26.1B parameters (fine-tuned from Qwen3.8-27B) and Flash v1.3 about 4.5B (distilled onto Qwen3.5-4B), both Apache 2.0; running v1.3 locally in BF16 needs about 49 GiB of GPU memory
  • New Locate feature: extracts verbatim spans by label or instruction with start and end character offsets and a score, covering NER, PII detection and extractive QA, and can share a request with classification and scoring questions
  • Classification (vendor-reported): 91.6% mean accuracy across 63 classification tasks, 6.54 points above Perplexity's production classifiers; ahead of Jev 1.14 in all 7 domains, by up to 12.9 points in finance
  • Retrieval and stability: nDCG@10 of 83.99 across 72 retrieval tasks, 1.81 above Qwen3-Reranker-8B; answers changed on only 0.27% of questions under input rewrites, versus 2.69% to 4.84% for Jev 1.14
  • Where to get it: Hugging Face downloads, the Perplexity Decisions API, and OpenRouter, where both v1.3 models are free until November 19
Read full story
Linked Models & Agents:
⚡ pplx-decider-v1.3⚡ pplx-decider-flash-v1.3⚡ Qwen3.8-27B⚡ Qwen3.5-4B⚡ Jev 1.14⚡ Qwen3-Reranker-8B🌐 Perplexity🌐 Hugging Face🌐 OpenRouter
#Open weights#Decision model#Classification#Span extraction#Routing#Apache 2.0#Perplexity
A
Artificial Analysis@ArtificialAnlys·
👀35
🔥84
View on 𝕏 ↗

Artificial Analysis Launches Decision Index v2: New Calibration-Aware Decision Score, 45% Private Data, Drex Max v1.1 Narrowly Leads Decision Models at 73

On November 8 (morning of November 9 in Singapore), independent evaluator Artificial Analysis released version 2 of its Decision Index for decision models. The headline metric is now a Decision Score that rewards both correct answers and well-calibrated confidence, the private-data share rises from 25% in v1 to 45%, and two new evaluations are added: AA-Paperclip with 1,500 private documents and AA-Symmetry for answer consistency. Among decision models, Nace AI's Drex Max v1.1 and Fastino Labs' GLiDE v1.2 both score 73, with Drex Max narrowly ahead, and Base Labs' Mammoth 27B scores 72. General LLMs score higher, with Claude Haiku 5.5 (Max) at 77 and GPT-6 Luna (Max) at 76, but cost $6,240 and $1,530 per 1M tasks versus $102 for Drex Max.

Artificial Analysis Launches Decision Index v2: New Calibration-Aware Decision Score, 45% Private Data, Drex Max v1.1 Narrowly Leads Decision Models at 73

Key Takeaways

  • New metric (official leaderboard): per-question score = 50 x (1 + c - (p - c)^2), equal overall to 50 x (1 + accuracy - Brier score); right answers always score above 50 and wrong ones below 50, with confidence spreading scores within each half
  • New weights: AA-Paperclip 30%, AA-Symmetry 15%, EvalSafe-O*NET-AA 15%, AA-Omniscience-Decision v1.1 15%, HLE-MultiChoice v1.1 15%, NVIDIA When2Call 10% (down from 25%); the saturated WildGuardTest was removed
  • Top decision models (vendor-run evaluation): Drex Max v1.1 73, GLiDE v1.2 73, Mammoth 27B 72; general LLMs Claude Haiku 5.5 (Max) 77 and GPT-6 Luna (Max) 76
  • Cost gap above 60x: Celeris-1.1 Decision $14 per 1M tasks, Drex Max v1.1 $102, Clef $859; GLiDE v1.2 bills a document's input for every question and costs $780, about 8x Drex Max
  • Speed: Celeris-1.1 Decision takes 56 ms per task and Elephant 9B 70 ms, versus 11.34 s for GPT-6 Luna (Max) and 32.65 s for Claude Haiku 5.5 (Max)
Read full story
Linked Models & Agents:
⚡ Drex Max V1 1⚡ Glide V1 2⚡ Mammoth 27b⚡ Elephant 9b⚡ Celeris 1.1 Decision⚡ Claude Haiku 5.5⚡ GPT-6 Luna🌐 Artificial Analysis🌐 Nace AI🌐 Fastino Labs🌐 Base Labs🌐 Anthropic🌐 OpenAI🌐 NVIDIA🌐 TypeSafe
#Decision Index#Artificial Analysis#decision models#Benchmark#Calibration#Brier score#AA-Paperclip#AA-Symmetry
S
Sarvam@SarvamAI·
👀37
🔥81
View on 𝕏 ↗

Sarvam Releases Prism, an Open-Weight Decision Model Built on Sarvam 105B That Returns Calibrated Option Probabilities in About 44 ms, Apache 2.0

Indian AI company Sarvam released Sarvam Prism on November 3, a "System One" decision model for enterprise workflows: given a context and questions with fixed answer options, it returns a probability for every option in a single forward pass instead of generating text. Sarvam says that across the 25 decision models it tested, Prism averages 71.8 over three indices with a median response time of 44 ms on one B200, the highest score under 100 ms, and scores 82.5 on its Indian-language benchmark. Weights are on Hugging Face under Apache 2.0, with a hosted API as well. All scores are Sarvam's own measurements.

Sarvam Releases Prism, an Open-Weight Decision Model Built on Sarvam 105B That Returns Calibrated Option Probabilities in About 44 ms, Apache 2.0

Key Takeaways

  • Trained on the Sarvam 105B (105B-10B MoE) base; it only outputs option probabilities, so the answer is always one of the given options.
  • Vendor-reported indices: Jev Decision Index 67.0, Sarvam's recreation of the Artificial Analysis decision index 65.9, internal Indian-language benchmark 82.5, average 71.8; Perplexity Decider v1.2 averages 68.0 at about 117 ms.
  • Calibration: on English answers at 90%+ confidence it expects 3.3% errors and makes 2.2%; in Indian languages 5.4% expected vs 4.4% actual, while rivals underestimate errors by 1.5x to 4.8x.
  • By sorting options into a fixed order before scoring, it changed 0 of 2,000 answers across four shuffled option orders.
  • On 700 recorded agent runs it judges correctly 78.5% of the time overall, vs 70.7% for Claude Haiku 5.5.
  • Self-hosting needs one FP8-capable GPU with about 120 GB free memory (such as B200 or H200) and about 106 GB of disk for weights.
Read full story
Linked Models & Agents:
⚡ Sarvam Prism⚡ Sarvam 105B⚡ Perplexity Decider v1.2⚡ Jev 1.14⚡ Microsoft-Decision-1⚡ Claude Haiku 5.5🌐 Sarvam
#Sarvam#Sarvam Prism#decision model#Calibration#open weights#Indian languages#enterprise AI
C
Cadence@Cadence·
👀55
🔥82
View on 𝕏 ↗

Cadence launches Maestro, a domain-specialized AI model portfolio for agentic chip and system design built in part on NVIDIA Nemotron, claiming up to 30% better results and 5x token efficiency

EDA leader Cadence announced the Maestro AI Model Portfolio on October 26: domain post-trained LLMs, trained decision models and self-improving agent skills that power its AuraStack, ChipStack, InnoStack and ViraStack AI Super Agents as they drive Cadence EDA and SDA tools. Cadence solvers score outputs against engineering constraints to close a reinforcement learning loop. Initial results per Cadence show up to 30% higher quality and up to 5x token efficiency versus a general-purpose foundation model. NVIDIA contributes Nemotron and OpenShell, and the stack runs on Cadence cloud, public cloud or on-premises. The first models reach select early access customers in Q4 2026.

Key Takeaways

  • Maestro is a portfolio, not one model: domain post-trained LLMs, decision models and self-improving agent skills, mixed with frontier and open-weight models per task
  • Deterministic solvers and EDA tools score outputs against engineering objectives and constraints, feeding a verified reinforcement learning loop
  • Per Cadence: up to 30% higher quality of results and up to 5x token efficiency versus a general-purpose foundation model; no third-party evaluation yet
  • NVIDIA supplies Nemotron models and the OpenShell sandbox reference architecture; customers can post-train on their own designs and run on-premises
  • First models go only to select early access customers in Q4 2026; Socionext expects up to 5x productivity from deploying InnoStack
Read full story
Linked Models & Agents:
⚡ Cadence Maestro⚡ NVIDIA Nemotron🤖 Cadence AI Super Agents🤖 ChipStack🤖 InnoStack🌐 Cadence🌐 NVIDIA🌐 Socionext🌐 Silicon Labs
#Cadence#Maestro#EDA#chip design#agentic AI#NVIDIA Nemotron#reinforcement learning#On premises
ADSponsored
B
Baseten@baseten·
👀119
🔥85
View on 𝕏 ↗

Baseten open-sources decision models mammoth-1 (27B) and elephant-1 (9B): per-option probabilities instead of text, #1 on the Jev Decision Index

On Oct 21, Baseten's research team Base Labs released open decision models mammoth-1 (27B) and elephant-1 (9B), full fine-tunes of Qwen3.5 that skip text generation and return per-option probabilities in one forward pass for routing, classification and scoring. Per Baseten, mammoth-1 tops the Jev Decision Index 0.4 at 71.24 and elephant-1 ties for #1 on the Artificial Analysis Decision Index. Weights are on Hugging Face and the Baseten API is free through Oct 27.

Baseten open-sources decision models mammoth-1 (27B) and elephant-1 (9B): per-option probabilities instead of text, #1 on the Jev Decision Index

Key Takeaways

  • Benchmarks (vendor): mammoth-1 scores 71.24 on Jev Decision Index 0.4, #1 of 208 decision models; 65.9 on AA Decision Index
  • elephant-1 (9B): 62.6 on Jev, best 9B; 65.2 on AA Decision Index, tied #1 with d1-ultra
  • Training: full fine-tunes of Qwen3.5-27B / 9B on 6.21M decision examples incl. abstention and safety data
  • Mechanism: block-bidirectional attention plus shared-prefix caching, one forward pass for many questions; up to ~10x faster than a reasoning model (vendor claim)
  • Try it: Hugging Face weights (MIT); Baseten API free until Oct 27 11:59 PM PT with zero data retention
Read full story
Linked Models & Agents:
⚡ Mammoth 1⚡ Elephant 1⚡ Qwen3.5-27B⚡ Qwen3.5-9B🌐 Baseten🌐 Base Labs
#Decision model#Open weights#Baseten#Classification#Routing
P
Perplexity Developers@perplexitydevs·
👀120
🔥96
View on 𝕏 ↗

Perplexity Releases Open-Weight Multimodal Decision Models pplx-decider-v1.2 (27B) and Flash (4B) on OpenRouter and Vercel AI Gateway

Perplexity Developers (@perplexitydevs) announced the launch of pplx-decider-v1.2 (27B) and pplx-decider-flash-v1.2 (4B) across OpenRouter and Vercel AI Gateway. Taking both text and images as input, these open-weight multimodal decision models return typed answers with calibrated probabilities at just $0.02 per million input tokens for the 27B checkpoint and half the cost for the 4B Flash variant.

Perplexity Releases Open-Weight Multimodal Decision Models pplx-decider-v1.2 (27B) and Flash (4B) on OpenRouter and Vercel AI Gateway

Key Takeaways

  • Dual open-weight multimodal decision models: launches 27B pplx-decider-v1.2 and 4B pplx-decider-flash-v1.2 taking text or images and returning typed answers with probabilities
  • Ultra-low $0.02/1M token pricing: 27B checkpoint costs $0.02 per million input tokens, while 4B Flash runs faster at half the price
  • Live on OpenRouter and Vercel AI Gateway: provides drop-in rolling aliases for low-latency agent routing and multimodal classification
Read full story
Linked Models & Agents:
⚡ pplx-decider-v1.2⚡ pplx-decider-flash-v1.2🤖 Tool-Routing Agents🤖 Multimodal Agents🌐 Perplexity🌐 OpenRouter🌐 Vercel
#Perplexity#pplx-decider-v1.2#Decision Models#Multimodal#OpenRouter#Vercel AI Gateway#Agent Routing
H
Hugging Face Daily Papers@HuggingFace·
👀118
🔥96

JevAgentHarness & AgentDecision-4B: Sharing Control Between 4B Decision Models and LLMs Makes Agents 2.5x Faster and 89% Cheaper

Standard agent harnesses invoke massive LLMs at every step, even though most steps reduce to discrete multiple-choice decisions such as selecting the next tool. Researchers from UIUC, Yale, and Princeton introduce JevAgentHarness, AgentDecisionBench, and the open AgentDecision-4B model. By calibrating per-choice confidence thresholds and training on post-mistake states, JevAgentHarness makes agents 81.1%–89.0% cheaper and up to 2.5x faster while solving 91.5% of out-of-distribution tasks.

JevAgentHarness & AgentDecision-4B: Sharing Control Between 4B Decision Models and LLMs Makes Agents 2.5x Faster and 89% Cheaper

Key Takeaways

  • Uncovers the static-to-agentic gap: builds AgentDecisionBench and shows uncalibrated tool delegation drops agent task success by 19.0 points
  • Per-choice thresholding & post-mistake training: introduces JevAgentHarness and open AgentDecision-4B trained on states reached after agent errors
  • 2.5x faster and up to 89% cheaper: slashes inference cost by 81.1%–89.0% while achieving 91.5% task resolution on out-of-distribution benchmarks
Read full story
Linked Models & Agents:
⚡ AgentDecision-4B⚡ Jev🤖 JevAgentHarness🤖 Tool-Using Agents🌐 UIUC🌐 Yale University🌐 Princeton University
#Agent Harness#JevAgentHarness#AgentDecision-4B#Tool Calling#Inference Acceleration#Cost Optimization
X
Xunzhuo@XunzhuoLiu·
👀279
🔥88
View on 𝕏 ↗

vLLM Semantic Router open-sources Decision 3.0: six multimodal decision models from 0.6B to 27B, #1 on Jev Decision Index in its own evaluation

On Oct 11 the vLLM Semantic Router (vllm-sr) team released the Decision 3.0 family (d3, d3-flash, d3-mini, d3-nano, d3-lite, d3-edge) on Hugging Face under Apache-2.0. Built on Qwen3.5/3.8, the models take text, images and video and return probabilities for choice, yes/no and score questions in one call without generating text. Vendor-reported: the 27B d3 scores 64.7 on Jev Decision Index 0.3.1, ahead of Perplexity Decider v1.1 (62.8) and Decision 2.0 (55.9), with a 55 ms median text latency on one AMD MI325X.

vLLM Semantic Router open-sources Decision 3.0: six multimodal decision models from 0.6B to 27B, #1 on Jev Decision Index in its own evaluation

Key Takeaways

  • Six sizes: d3-edge 0.59B, d3-lite 0.85B, d3-nano 2.2B, d3-mini 4.5B, d3-flash 8.4B, d3 26.1B, all Apache-2.0 (HF collection)
  • Vendor-reported Jev Decision Index 0.3.1: d3 64.7 vs Perplexity Decider v1.1 62.8, Jev 60.1, Decision 2.0 55.9
  • Vision Index 0.3.1: d3 71.6 vs Perplexity Decider v1.1 70.6 and JEV-27B-VL 69.6
  • Median latency on one AMD MI325X, batch 1: d3 55 ms text / 273 ms image / 553 ms 10-second video; d3-edge 6.7 ms text
  • No text generation: every option gets a probability, and several Choice / Yes-No / Score questions share one call
Read full story
Linked Models & Agents:
⚡ Decision 3.0⚡ d3⚡ D3 Flash⚡ D3 Edge⚡ Qwen3.5⚡ Qwen3.8🌐 vLLM🌐 vLLM Semantic Router
#Open-source Models#Jev#Decision Models#vLLM
L
Liquid AI@liquidai·
👀160
🔥88
View on 𝕏 ↗

Liquid AI open-sources Open d1: d1-3B and d1-omni-600M decision models answer in one forward pass, 8 ms on RTX 4090

Liquid AI released two open-weight multimodal decision models: d1-3B (text+vision) and the experimental d1-omni-600M (text+image or text+audio). Instead of generating tokens they answer in a single forward pass. d1-3B scores 48.57 on Decision Index v0.2.1 (public split), first among sub-10B models and on par with the 12x larger Decider 35B-A3B, and answers in 8 ms on an RTX 4090 and 50 ms on a Jetson Orin Nano, with day-one llama.cpp support.

Liquid AI open-sources Open d1: d1-3B and d1-omni-600M decision models answer in one forward pass, 8 ms on RTX 4090

Key Takeaways

  • d1-3B: 48.57 on Decision Index v0.2.1 public split, #1 under 10B, on par with Decider 35B-A3B; d1-omni-600M: 15.95
  • Text benchmark mean: d1-3B 82.9 vs Decider 4B 81.1; d1-omni-600M 78.4 vs Decider 2B 77.1 (Civil Comments 95.8)
  • Single-question latency: 8 ms RTX 4090, 9 ms MI325X, 16 ms Jetson AGX Thor, 30 ms Apple M5 Pro, 50 ms Jetson Orin Nano
  • Packed throughput: 475 states/s on RTX 4090, 1,106/s on MI325X; three questions cost ~1.3x one
  • Weights on Hugging Face (incl. GGUF and w8a8), LFM Open License v1.0, day-one llama.cpp and NVFP4 support
Read full story
Linked Models & Agents:
⚡ d1-3B⚡ d1-omni-600M⚡ LFM2.5-VL-3B🌐 Liquid AI🌐 NVIDIA
#Liquid AI#Open-source Models#Decision Models#Jetson#llama.cpp
X
Xuan-Son Nguyen / ggml-org@ngxson·
👀196
🔥96

llama.cpp ships Decision Models: /v1/systemone scores options in one forward pass (Julia-1, Laya, Kev-4B, lev, OpenJev)

On 2026-10-02 ggml-org announced llama.cpp server support for Decision Models via /v1/systemone (PR #29818, merged 2026-10-02T09:56Z). Send a state plus typed questions; get option probabilities in one forward pass. Day-0 models: Julia-1, Laya, Kev-4B, lev, OpenJev (vision). Cloudflare Clef is planned next.

llama.cpp ships Decision Models: /v1/systemone scores options in one forward pass (Julia-1, Laya, Kev-4B, lev, OpenJev)

Key Takeaways

  • Official post 2026-10-02 + merged PR #29818 (2026-10-02T09:56Z, +2139/−15)
  • Endpoint POST /v1/systemone with choice/score/noul; probabilities + confidence; output_tokens=0
  • Day-0 GGUFs: Julia-1 / Laya / Kev-4B / lev / OpenJev; median ~3–43 ms/question on RTX PRO 6000 (blog table)
  • Distinct from Ollama SystemOne & Cloudflare Clef hosting; Clef listed as next in llama.cpp
  • Start: llama serve -hf ggml-org/Kev-4B-GGUF then curl /v1/systemone
Read full story
Linked Models & Agents:
⚡ Julia 1⚡ laya⚡ Kev 4b⚡ lev⚡ openjev🌐 Ggml org🌐 Huggingface
#llama.cpp#Decision Models#System One#Systemone#Jev#GGUF#local inference#agent routing#Kev-4B#OpenJev#Ggml
ADSponsored
C
Cloudflare@Cloudflare·
👀145
🔥97
View on 𝕏 ↗

Cloudflare open-sources Clef & Clef-flash decision models on Workers AI — 209ms / 38.8ms median

@Cloudflare (2026-10-01, #BirthdayWeek) launched its first in-house decision models Clef (27B) and Clef-flash (9B): hosted on Workers AI, Apache 2.0 weights, Jev System One–compatible; official blog cites ~2.5× / 13× median latency vs Jev and 2.2s vs 4.7s in a threat-intel Browser Run workflow.

Cloudflare open-sources Clef & Clef-flash decision models on Workers AI — 209ms / 38.8ms median

Key Takeaways

  • Primary: Cloudflare 2026-10-01 → Clef blog
  • Median latency (43 runs): Clef 209.3 ms, Clef-flash 38.8 ms vs Jev 524.1 ms (~2.5× / 13×)
  • Quality: BFCL 98.47/98.76; BANKING77 94.20; CLINC150+OOS 97.43; Clef family tops 7/10 decision benches
  • Shape: Qwen backbone + non-AR schema scoring; 64K ctx; vision; @cf/cloudflare/clef / clef-flash; $0.24 / $0.09 per M input tokens
  • Ship: Workers AI binding/REST; weights on HF; RL fine-tune via design-partner form
Read full story
Linked Models & Agents:
⚡ Cloudflare Clef⚡ Cloudflare Clef Flash🌐 Cloudflare
#Cloudflare#Clef#Clef-flash#Workers AI#decision model#Jev#BirthdayWeek#RL fine-tuning#Apache 2.0
O
Ollama@ollama·
👀78
🔥90

Ollama Python 0.6.3: native systemone() for Decision Models; think accepts model-defined levels

Official Ollama Python client v0.6.3 shipped 2026-09-29: adds ollama.systemone() matching server v0.35.0 /v1/systemone Decision Models (choice/noul/score); relaxes think to string levels; fixes multi-tool examples and case-insensitive image extensions. Requires Ollama ≥0.35.0 plus local decision models such as nimble/tev1.

Ollama Python 0.6.3: native systemone() for Decision Models; think accepts model-defined levels

Key Takeaways

  • Release v0.6.3 on 2026-09-29; ~6 commits vs v0.6.2 including examples/systemone.py
  • New ollama.systemone(model, state, questions) batches choice/noul/score Decision Model queries
  • Needs Ollama server v0.35.0+ and a local decision model (nimble/tev1)
  • think accepts string levels for model-defined thinking (#697/#744)
  • Install: pip install -U ollama==0.6.3
Read full story
Linked Models & Agents:
⚡ nimble⚡ tev1🤖 ollama🌐 Ollama
#Ollama#Python SDK#Decision models#Systemone#Release
O
Ollama@ollama·
👀151
🔥93

Ollama 0.35.0: Decision Models via /v1/systemone (Jev); Nimble & Tev1 return choices, probabilities, confidence

Ollama shipped v0.35.0 on 2026-09-28 (GitHub prerelease): Decision Models via TypeSafe’s Jev API at /v1/systemone. Instead of free text, models answer choice/noul/score questions with options, probabilities, and confidence—for ticket triage, model routing, and classification. First models: Nimble (Bespoke Labs) and Tev1 (Together AI). ~7 commits / 66 files vs v0.34.4; Settings no longer blocks on discovery; macOS update menu and stalled MLX downloads fixed.

Ollama 0.35.0: Decision Models via /v1/systemone (Jev); Nimble & Tev1 return choices, probabilities, confidence

Key Takeaways

  • v0.35.0 (prerelease) adds Decision Models at POST /v1/systemone (TypeSafe Jev)
  • Question types: choice (with probabilities), noul, score — not free-text generation
  • First models: Nimble (Bespoke Labs) and Tev1 (Together AI) on ollama.com/library
  • ~7 commits / 66 files vs v0.34.4; Settings/macOS update/MLX download fixes
Read full story
Linked Models & Agents:
⚡ Nimble⚡ Tev1⚡ Jev🤖 Decision Models🤖 model routing🌐 Ollama🌐 TypeSafe🌐 Bespoke Labs🌐 Together AI
#Ollama#Decision Models#Systemone#Jev#Nimble#Tev1#Classification#Routing
U
Unsloth@unslothai·
👀107
🔥94

Unsloth v0.1.900-beta: local Laya Decision models + Skills editor; ~4.5× LTX-2.3 clips and up to 6.3× faster VAE decode

Unsloth shipped v0.1.900-beta on 2026-09-28 (PyPI unsloth 2026.9.12): Desktop runs Laya Decision models locally via a Jev-compatible /v1/systemone endpoint, adds a Skills CRUD editor and Library viewers for PDF/Office, and improves Apple Silicon with batched serving, structured outputs, and TurboQuant KV. Media pipelines claim ~4.5× LTX-2.3 clips, 1.7–6.3× VAE decode, and up to ~1 minute faster MiniMax-H3 first render with 25–29 GiB lower peak VRAM.

Unsloth v0.1.900-beta: local Laya Decision models + Skills editor; ~4.5× LTX-2.3 clips and up to 6.3× faster VAE decode

Key Takeaways

  • Shipped v0.1.900-beta — local Laya Decision API + Skills editor + Library viewers
  • LTX-2.3 ~4.5×; VAE decode 1.7–6.3×; MiniMax-H3 first render up to ~1 min faster, −25–29 GiB peak VRAM
  • Apple Silicon: batched serving, structured outputs, TurboQuant KV; ModelScope downloads
  • Docs: docs.unsloth.ai
Read full story
Linked Models & Agents:
⚡ Laya⚡ LTX-2.3⚡ MiniMax-H3⚡ Gemma-4🤖 Unsloth-Studio🌐 Unsloth
#Unsloth#Decision Model#Jev#Skills#Apple Silicon#MLX#LTX#Fine tuning
O
OpenClaw@openclaw·
👀125
🔥92

OpenClaw 2026.9.6: managed updates, restart recovery, 30-day Usage, plus Opus 5.5 / GPT-6 Sol·Luna / Grok 4.7

OpenClaw shipped v2026.9.6 on 2026-09-23 (macOS DMG rebuilt 2026-09-24 after launch crash #156861): clearer managed-update outcomes, restart recovery for unfinished work, complete 30-day Usage reporting, a GitHub reader, remote workspace Files/Memory/Skills, live meeting notes, optional TypeSafe Jev Decision Models, and chat models Claude Opus 5.5, GPT-6 Sol/Luna, and Grok 4.7. Scale: 2,614 PRs, 178 direct commits, ~350 contributors.

OpenClaw 2026.9.6: managed updates, restart recovery, 30-day Usage, plus Opus 5.5 / GPT-6 Sol·Luna / Grok 4.7

Key Takeaways

  • Verified scale on docs: 2,614 PRs / 178 commits / ~350 contributors
  • Managed updates + restart recovery; Usage covers a full 30-day window
  • GitHub reader; remote workspaces gain Files/Memory/Skills; live meeting notes
  • Decision Models (TypeSafe Jev/local) + Opus 5.5, GPT-6 Sol/Luna, Grok 4.7
  • Install [email protected]; macOS users hit by #156861 should reinstall the rebuilt DMG (#156881)
Read full story
Linked Models & Agents:
⚡ Claude Opus 5.5⚡ GPT-6 Sol⚡ GPT-6 Luna⚡ Grok 4.7⚡ TypeSafe Jev🤖 OpenClaw🌐 OpenClaw
#OpenClaw#Agent#Usage#restart recovery#Opus 5.5#GPT-6#Grok 4.7
You've caught up with all stories