AI Coding Live News

On Oct 27 OpenBMB (ModelBest) open-sourced the MiniCPM-OCR suite with two models: MiniCPM-OCR 1.0, a 1.3B-parameter document parser that scores 93.68 on Real5-OmniDocBench and 54.7 on OCRBench v2 (vendor-reported), and MiniCPM-OCR Jev, a 2B OCR quality judge that scores any parser's output region by region from 0 to 1 and flags content that needs human review. Code and weights are Apache 2.0, with day-one support for vLLM, SGLang, llama.cpp, Ollama and multiple chip platforms.

Zoom

On Oct 28 DiffuSpace, whose members come from the HKU NLP group behind Dream 7B, released its first model, Clover-35B-A3B-Preview: a mixture-of-experts diffusion language model initialized from Qwen3.6-35B-A3B, with 35B total and about 3B active parameters, a 128K context window and block-parallel denoising. Vendor-reported results include 73.6 on SWE-bench Verified, 51.7 on Terminal-Bench 2.1 and 83.8 on GPQA Diamond, with 680 tokens/s on a single GB300 in agent workloads. Weights are Apache 2.0 on Hugging Face, ModelScope and GitHub, with SGLang and llama.cpp support.

Zoom

Recursive, founded by Richard Socher, has officially open-sourced Flint (RecursiveSI/flint), a 39.8B Mixture-of-Experts (MoE) coding agent model under Apache 2.0 with day-zero Ollama support (ollama run flint). Activating just 2.5B parameters per token across 47 MoE layers with a 262K context window, Flint hits 125+ tokens/sec on an M4 Pro MacBook Pro. It sets new benchmarks for sub-50B local models, scoring 74.2% on SWE-bench Verified and 89.7% on SWE-bench Pro v2.

Zoom

On Oct 19 (UTC) Unsloth released dynamic GGUF quants and a run-locally guide for DeepSeek-V4.1-Flash-1017. Per Unsloth, the 1-bit UD-IQ1_M quant is 250.4 GB (vs 508.4 GB for lossless Q8), keeps 88.2% top-1 agreement with the official weights on long17k, and runs on 128 GB+ RAM machines, because the model's 110.6–202.8 GB Engram n-gram table can stay on SSD via mmap. Running it currently requires Unsloth's llama.cpp PR build or Unsloth Desktop.

Zoom

Nous Research shipped Hermes Agent v0.22.0, "The Threshold Release", on Oct 14 (UTC). Fresh desktop installs open a guided setup chat on a free trial model (nous/welcome) with no API key, connectors reach Gmail, Linear and Notion through the Nous tool gateway, a one-click llama.cpp runtime runs local models, installs now follow tested SemVer stable releases, and desktop builds are signed (including the Microsoft Store). The source tree shrank from about 1.06M to 698K lines (−34%).

Zoom

On 2026-10-12 OpenBMB released MiniCPM-V 4.7, an Apache-2.0 open multimodal family with a 35B-A3B variant (the series' first MoE) and a 1B variant. Its Lookahead Expert Scheduling (LES) prefetches experts across layers and, combined with layer-wise mixed-precision quantization and DSpark speculative sampling, lifts SSD-offloaded decoding on an iPad Pro M4 (16GB) from 2.97 to 17.39 tokens/s (5.86x), per OpenBMB. Weights are on Hugging Face, ModelScope and Ollama, with vLLM, SGLang, llama.cpp and iOS/Android/HarmonyOS support.

Zoom

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.

Zoom

Large language model serving exhibits bifurcated hardware constraints: prefill is compute-bound, benefiting from high-throughput low-precision tensor operations (e.g., NVFP4), while decode is memory-bandwidth-bound, requiring ultra-compact weight storage (e.g., 2-3 bit weight-only) to minimize token-generation memory traffic. Uniform quantization forces suboptimal compromises across both phases. Researchers from IST Austria introduce Disaggregated Quantization (DQ, arXiv:2609.26333), decoupling computation formats, quantization precision, and physical memory placement across prefill and decode. Paired with Offloaded Disaggregated Prefill (ODP) to stream NVFP4 prefill weights asynchronously from NVMe SSDs, DQ delivers a 1.78x time-to-first-token (TTFT) speedup on Qwen3.8-27B at 8K prompt length within llama.cpp, while supporting disaggregated serving in vLLM on architectures up to 2.8T parameters.

Zoom

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.

Zoom

llama.cpp v0.5.0 (2026-09-23) accelerates CUDA conv2d via implicit GEMM, adds Metal MoE and SSM_CONV fusion, lets the server bind multi-address/--host lists and UNIX sockets, and adds HRM-Text (DFM Mimir 1B), MiMo-V2.6, and HunyuanOCR (DFlash) conversion plus ggml 0.25.0 backends.

Zoom

Local LLM inference runtime Ollama (@ollama) officially rolled out v0.40.0. The standout architectural milestone is the default adoption of Apple's native MLX framework on all Apple Silicon Mac devices (M1 through M5), replacing the legacy CPU/Metal compute paths. Supporting major architectures like Qwen 3.8 and Llama 3 natively, the update harnesses Apple's unified memory bandwidth, slashing time-to-first-token by 42% and nearly doubling continuous generation throughput without manual configuration.

Zoom

Cline VS Code extension shipped v4.1.21 on 2026-09-24 with a new Japan OpenAI-compatible provider (ai&), a refreshed catalog of 6,386 models across 209 providers (19 unpinned defaults change—11 to Claude Opus 5.5 including GitHub Copilot and Vertex), and compact-and-retry once when local servers (llama.cpp/Ollama/LM Studio) hit output-token limits. Minimum js-yaml raised to 4.3.2 for frontmatter parsing security.

Zoom