Loading…
Topic
Open Source
561 posts about Open Source. Every summary links to the original.
- 561 posts
- 36 companies
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Newest first
Posts
Shipping a Trillion Parameters With a Hub Bucket: Delta Weight Sync in TRL
Async RL weight synchronization currently requires the trainer to send a complete model to the inference engine after each optimizer step, creating a critical-path transfer that can reach roughly a terabyte for a frontier 1T-parameter checkpoint. The article attributes the opportunity for reduction to bf16 arithmetic: at RL learning rates, many updates fall below the bf16 visibility threshold, leaving about 99% of weights bit-identical between steps and more than 98% even in the reported worst case. TRL's delta-weight-sync implementation records changed elements around the optimizer step, encodes them in sparse safetensors, uploads each delta to a Hugging Face Bucket, and lets vLLM fetch and apply it. On Qwen3-0.6B, this reduces each per-step payload from 1.2 GB to 20–35 MB, while a demonstrated setup connected separate trainer, vLLM, and Wordle Spaces through one bucket. Open issues include extra CPU bf16 snapshots, fixed anchor cadence, and unmeasured multi-node FSDP2 support.
5 countries, 12 days, one ClickHouse: inside Alexey's APJ AI Tour
ClickHouse CTO and creator Alexey Milovidov completed a 12-day APJ tour spanning India, Singapore, Indonesia, South Korea, and Japan, with customer meetings, industry events, executive sessions, press interviews, and community meetups. The itinerary covered six cities and included three Confluent Data Streaming World Tour events, four ClickHouse meetups, and more than 10 enterprise customer meetings across varied industries and workloads. Discussions centered on real-time analytics, connected products, and analytics at industrial scale, while in-person conversations supplied feedback that the post says helps improve ClickHouse. The trip concluded with the ClickHouse Japan Launch Event on April 27, marking the company’s official entry into that market. The account presents physical presence as a way to extend ClickHouse’s open-building culture beyond GitHub and strengthen its feedback loop with APJ customers and communities.
Introducing CostBench: an open benchmark for data warehouse cost-performance
Traditional data warehouse benchmarks measure query execution speed without accounting for the varying compute costs across platforms. CostBench introduces an open benchmarking framework designed to evaluate data warehouses on a performance-per-dollar metric across both read and write workloads. Its initial release evaluates ClickHouse Cloud, Snowflake, Databricks, BigQuery, and Redshift using 43 production-derived analytical queries on an anonymized dataset combined with each vendor's compute billing model. In these read-side tests, ClickHouse Cloud maintained high speed at lower costs as data scaled, outperforming the nearest competitor by 23 times in cost-performance. The benchmark repository openly publishes its scripts, configurations, pricing assumptions, and raw JSON results to ensure all comparative metrics remain fully inspectable and reproducible.
Integrating the Rust Delta Kernel into ClickHouse
ClickHouse initially built native protocol support to read Delta Lake tables, but maintaining pace with the format's evolving specification created high complexity and fragmented features. To resolve this, engineers integrated the Rust Delta Kernel to offload protocol handling while retaining ClickHouse's optimized Parquet data reading. The Delta Kernel parses JSON transaction logs, resolves snapshots, manages transactional write metadata, and exposes schema evolution mapping alongside Change Data Feed streams. ClickHouse also contributed upstream enhancements to the Rust kernel, adding dynamic runtime logging and asynchronous metadata processing via FFI handles. While ClickHouse can attach to existing Delta tables, the integration cannot yet initialize empty Delta Lake tables.
OlmoEarth v1.1: A more efficient family of Earth observation models
OlmoEarth v1.1 is a new family of Earth observation models designed to reduce the compute required for large-scale satellite imagery predictions. Compute dominates the lifecycle from data export through preprocessing, inference, and post-processing, motivating a focus on transformer token sequence length. For Sentinel-2, the models collapse three resolution-specific tokens into one, reducing token counts threefold; this initially caused a 10 ppt drop on the m-eurosat kNN benchmark, so the pre-training regimen was modified. OlmoEarth v1.1 runs up to three times cheaper than v1 while maintaining similar performance across a mix of research benchmarks and partner-built tasks, although some regressions remain, and the same dataset is used to isolate methodological effects.
What's new in ClickStack - April 2026
ClickStack's April release introduces several core enhancements across querying, alerting, and dashboard visualization capabilities. The platform added SQL-powered alerting, allowing teams to express complex statistical rules like rolling baselines and anomaly detection using native ClickHouse SQL and window functions without external pipelines. To improve search and filtering performance across common observability workloads, the default otel_logs schema was redesigned with ClickHouse text indexes and adjusted ordering keys. Furthermore, heatmaps transitioned into a first-class chart type inside the shared visualization system for latency distribution tracking. The update also delivers persistent alert execution error reporting, expanded threshold comparison operators, and per-series number formatting across dashboards.
Introducing the Ettin Reranker Family
Tom Aarsen introduces six Sentence Transformers CrossEncoder rerankers, ranging from 17M to 1B parameters, alongside their training data and full recipe. Each model uses an Ettin ModernBERT encoder with unpadded attention, RoPE, GeGLU, up to 8,192 tokens of context, and a modular classification head whose ablations favored CLS pooling over mean pooling. Training applies pointwise MSE distillation to scores from mixedbread-ai/mxbai-rerank-large-v2 using a broad and retrieval-focused dataset mixture. The conclusion reports state-of-the-art results at every released size, with the 400M model within 0.0024 and the 1B model within 0.0001 of the 1.54B teacher on MTEB, while all models outperform the ms-marco-MiniLM-L*-v2 family on MTEB and NanoBEIR.
Granite Embedding Multilingual R2: Open Apache 2.0 Multilingual Embeddings with 32K Context — Best Sub-100M Retrieval Quality
Granite Embedding Multilingual R2 introduces two Apache 2.0 multilingual embedding models built on ModernBERT: a 97M-parameter compact model and a 311M full-size model. Both cover more than 200 languages, receive enhanced retrieval training for 52 languages and nine programming languages, and support context lengths up to 32,768 tokens, compared with 512 for the R1 models. The compact model scores 60.3 on MTEB Multilingual Retrieval, while the full-size model scores 65.2; the reported results place them among the strongest open models in their parameter ranges. R2 replaces the XLM-RoBERTa architecture with alternating attention lengths, rotary position embeddings, newer multilingual and code tokenizers, and Flash Attention 2.0 support, while the 311M model adds Matryoshka embeddings and both models provide ONNX and OpenVINO weights.
Faster Queries and New Capabilities with the Open-Source Databricks JDBC Driver
The open-source Databricks JDBC driver 3.x and later releases address connectivity needs affecting dashboards, spreadsheets, applications, and other data workflows. Compared with the legacy 2.x driver, they deliver up to 30% faster retrieval of large query results while preserving backward compatibility. Architectural changes add full Arrow-based transfer for JDK 16+, a first-class asynchronous execution API, stream-based ingestion into Databricks Volumes, and integration with the Statement Execution API. The driver also supports stored procedures, multi-statement transactions, Unity Catalog metric views, query tags, geospatial and complex data types, plus built-in client telemetry for query latency, metrics, and errors. Its open-source ownership and Databricks maintenance are presented as enabling faster fixes, new features, transparency, contributions, and tighter platform integration.
ClickHouse Release 26.4
ClickHouse Release 26.4 introduces 39 new features, 45 performance optimizations, and 238 bug fixes, focusing on broader SQL compatibility and indexing improvements. The update expands standard SQL support by allowing VALUES clauses as table expressions with column aliases, PostgreSQL-style EXTRACT units, NATURAL JOIN, and SET TIME ZONE commands. For text search, LIKE and ILIKE pattern queries using splitByNonAlpha tokenizers scan inverted index dictionaries instead of executing full-table scans when the dedicated evaluation setting is active. The release also provides the JSONAllValues function to extract all leaf values from a JSON column into a string array, enabling text skip indexes that significantly reduce scanned data and improve query latency.
Introducing Postgres Query Insights in ClickHouse Cloud
ClickHouse Cloud introduced Query Insights in preview for Managed Postgres to help teams diagnose slow database queries. Powered by the open-source extension pg_stat_ch, the tool streams per-statement telemetry directly into ClickHouse to enable fast aggregations across billions of execution rows. Postgres normalizes query literals into placeholders during the parse-analyze phase before telemetry leaves the database, preventing sensitive data exposure. Telemetry emission maintains around three percent producer overhead by utilizing a non-blocking try-lock on a shared-memory ring buffer. Users can examine high-level database health metrics, rank query patterns by runtime or latency percentiles, and inspect execution flyouts detailing disk spills, cache hits, and parallel worker behavior.
Fixing JavaScript observability, one library at a time
JavaScript application performance monitoring tools rely on monkey-patching via require-in-the-middle and import-in-the-middle, which breaks with ECMAScript Modules, bundlers, and non-Node runtimes. To solve this, an initiative is replacing monkey-patching with Node's built-in diagnostics_channel TracingChannel API across server-side JavaScript libraries. Under this pattern, libraries publish structured events while monitoring vendors subscribe to them with zero overhead when unlistened. The author leveraged Claude Code workflows alongside direct maintainer communication to draft proposals, implement code, and manage reviews across 44 target packages. Ten libraries including mysql2, node-redis, ioredis, and unjs modules have merged support, while active efforts continue on shared OpenTelemetry mapper registries and remaining ecosystem packages.
Adding Benchmaxxer Repellant to the Open ASR Leaderboard
The Open ASR Leaderboard is adding private English speech-recognition datasets from Appen Inc. and DataoceanAI to evaluate scripted and conversational speech across several accents. To reduce benchmark-specific optimization (“benchmaxxing”) and test-set contamination, the datasets remain private, while the default Average WER continues to use public datasets only. A toggle exposes the private data, with macroaverages for overall, scripted, conversational, American, and non-American subsets; individual split scores are withheld to discourage provider- or accent-specific optimization. The leaderboard uses a shared Hub dataset, normalized transcripts and outputs, and open-source evaluation code, while private evaluation is run after public-set verification when models are submitted.
Welcoming Diel Duarte, our new Open source Engineer
Resend introduces Diel Duarte as its new open source engineer, bringing more than 13 years of experience building web applications and SDKs. His focus is developer experience, especially API design with clean ergonomics that lets developers concentrate on building rather than wrestling with interfaces. Diel describes a path from studying programming and computer fundamentals at Cotemig to following Resend as a user and joining after an interview process he found unusually human; he cites Resend’s DX and craft, along with Zeno’s open source work, as major references. Outside work, he draws inspiration from deliberately curated feeds, new languages and APIs, visual-design communities, and creator friends, while keeping music and interior design as interests; he names Warp Terminal as his favorite tool and Japan as a memorable destination.
Granite 4.1 LLMs: How They’re Built
Granite 4.1 is a family of dense decoder-only language models in 3B, 8B, and 30B sizes, trained on approximately 15T tokens. Its five-phase pre-training pipeline moves from broad web-scale data to progressively curated mixtures, increases math and code emphasis, and extends context from 4K to 512K tokens through staged long-context training. Supervised fine-tuning uses about 4.1M curated samples evaluated with LLM-as-Judge and rule-based checks, while reinforcement learning uses on-policy GRPO with DAPO loss. The 8B instruct model matches or surpasses the earlier Granite 4.0-H-Small 32B-A9B MoE model despite its simpler dense architecture and fewer parameters, and the Granite 4.1 models are released under Apache 2.0.
From init to deploy: building with agents and Stripe Projects
The post describes an internal agentic transcription app built to reduce the manual effort required to correct auto-generated subtitles for Stripe Developers videos. The app uses a Next.js interface with transcript, chat, and corrected-output panels, while Stripe Projects CLI links Vercel and OpenRouter and populates required environment variables. An LLM client supports selectable models, iterative instructions, and a transcription agent whose system prompt emphasizes preserving transcript structure; the agent can also use Stripe documentation search through MCP. In a benchmark on a 771-line meetup script, Gemini 2.5 Pro reached 99% accuracy, while the tested models reduced editing from more than 30 minutes to just over two minutes of generated edits plus a human check. The reported overall reduction was 4x, with testing costs totaling $2.54.
Comparing ClickHouse versions with clickhousectl
Evaluating query performance differences across ClickHouse versions previously required managing Docker containers or manually downloading historical binaries from GitHub releases. The clickhousectl command-line interface simplifies this process by downloading specific ClickHouse binaries, running multiple background server instances concurrently with automatic port assignment, and executing queries across local environments. Users can copy table schemas between local instances by piping raw output from SHOW CREATE TABLE queries and replicate datasets using ClickHouse's remote table function. In a benchmark comparing ClickHouse 25.12 and 26.3 on an S3-backed Parquet dataset, version 26.3 introduced a Parquet metadata cache that reduced repeat query times from roughly 9 seconds to 1–2 seconds. System tables like system.server_settings and system.metrics enable operators to inspect cache configurations, eviction policies, and memory consumption directly.
Introducing NVIDIA Nemotron 3 Nano Omni: Long-Context Multimodal Intelligence for Documents, Audio and Video Agents
NVIDIA Nemotron 3 Nano Omni is introduced as an omni-modal understanding model for document analysis, multiple-image reasoning, automatic speech recognition, long audio-video understanding, agentic computer use, and general reasoning. It extends the Nemotron multimodal line from vision-language processing to text, image, video, and audio, targeting long, dense inputs such as 100-plus-page documents and mixed audio-visual evidence. Its unified encoder-projector-decoder combines a Nemotron 3 Nano 30B-A3B hybrid Mamba-Transformer MoE backbone with C-RADIOv4-H and Parakeet-TDT-0.6B-v2 encoders; dynamic visual resolution, Conv3D tubelets, and Efficient Video Sampling support multimodal context compression. The post reports leading or top results across document, video, audio-video, voice, and ASR benchmarks, alongside up to 9x higher throughput and 2.9x single-stream reasoning speed on multimodal use cases, and provides BF16, FP8, and NVFP4 checkpoints.
Introducing the OSSCAR Index
Supabase has launched the OSSCAR Index, a quarterly ranking of fastest-growing open source organizations, with the site, data, and scoring code released as open source. It addresses the limits of rankings based on accumulated stars, downloads, or contributors by measuring community growth within two divisions: Emerging organizations below 1,000 GitHub stars at quarter-start and Scaling organizations at or above that threshold. Scores combine normalized net new stars, unique contributors, and npm, PyPI, and Cargo downloads using an L² norm, with minimum thresholds and no penalty for missing or declining signals. The Q1 2026 edition identifies Openclaw as a breakout project, places AI-agent projects prominently in Emerging, and reports global representation; quarterly updates, expanded package coverage, an RFC process, and historical rankings are planned.
Formatting an entire 25 million line codebase overnight: the rubyfmt story
Stripe’s 25-million-line Ruby codebase needed a true autoformatter; before rubyfmt, no industry-wide Ruby formatter existed, and earlier tools had crashed on its files. rubyfmt began as a zero-configuration open-source project motivated by formatting disputes, Ruby’s slow startup, and the uncertainty faced by engineers arriving from other languages. To meet a 100ms target, it initially used a hand-authored C launcher and later moved formatting into Rust, with ripper parse trees first passing through JSON and eventually being deserialized directly from Ruby VALUE objects. Stripe adopted per-file opt-in, expanded coverage cautiously, and formatted the codebase on a Saturday in 2024; the excerpt reports that all 42-million-line Ruby codebase is now formatted. In 2025, migrating from ripper to Prism removed the need to link a Ruby VM and made rubyfmt smaller and dramatically faster.
Related topics
All topics →Weekly digest
The best engineering reads, once a week.