AI for biology has several evidence levels. Keep them separate
Biological agents can turn existing measurements into useful leads. A map of the evidence shows where interpretation ends and validation begins.
AI writing, AI reviewing and the verification bottleneck
Citation matching, technical criticism, executable checks and formal proofs answer different questions. A map of what AI review can establish.
From fuzz tests to proofs: inside an AI verification workflow
A maintainer’s guide to the artifacts behind AI-written proofs, from fuzz targets and binary hashes to assumptions and release changes.
What the AISI incident reveals about evaluation boundaries
AISI’s incident separates task success, permitted action and containment. Its follow-up experiments explain why each needs its own evidence.
Apertus 1.5 broadens the public-model research stack
Apertus adds image and audio input, with public tooling that exposes the work between a text checkpoint and a multimodal assistant.
Coding-agent productivity needs a systems denominator
Coding-agent activity is easy to count. Useful delivery requires accounting for review, infrastructure, retries and the quality of accepted changes.
FlashProve and the limits of a cryptographic AI receipt
FlashProve makes model execution a checkable claim. Its research prototype also exposes the limits of what an AI receipt can guarantee.
Inside Google’s building blocks for agent-assisted wet labs
Google’s lab studies separate physical execution, molecular search and experimental validation. The distinctions explain what the agents achieved.
GPT-6.1 Terra brings a specialised pitch to computer work
Terra targets visual reasoning and screen-based tasks. Its release makes the model, action environment and saved-result checks separate questions.
Model merging versus distillation: what the new evidence compares
Merging changes parameters; distillation trains on behavior. P²-MOPD explores a hybrid, with evidence bounded by its models, teachers and training budget.
Muse Spark Open and Apertus 1.5 expose different kinds of openness
Weights, training evidence, licence terms and reproducible evaluations are separate criteria. Two releases make the differences concrete.
The sleep framework puts forgetting at the centre of AI learning
Sleep consolidates temporary information into model parameters. Its experiments clarify what retention means and what the extra computation buys.
TorchNeuron’s public beta: what PyTorch teams should verify
TorchNeuron makes Trainium a native PyTorch device. Operator fallbacks, static shapes and numerical checks still decide migration readiness.
World models, video generators and simulators serve different jobs
Interactive generation, cinematic video and physics simulation expose different controls and evidence. Here is how to match them to a task.
Fact check: LiteLLM's Rust benchmark claims, tested against its own repo files
LiteLLM's homepage and benchmark blog credit its Rust gateway with about 7x lower overhead, 9x less memory and about 4.5x more requests per dollar. Fifteen claims, set against the CSV and JSON files LiteLLM published, with neutral verdicts, including the ones that check out.
LiteLLM's Rust gateway benchmark, checked against LiteLLM's own data
LiteLLM says its Rust gateway has about 7x lower overhead, 9x less memory and 4.5x more requests per dollar than the next-closest gateway. LiteLLM's own benchmark files show a latency and throughput near-tie, a mislabeled widest-pair memory chart, a modeled cost and a config gap, plus Rust leads on CPU and memory.
LiteLLM vs Bifrost latency: tracing the 7x overhead figure
LiteLLM says its Rust gateway has roughly 7x lower overhead than Bifrost. The concurrency sweep in LiteLLM's own benchmark repo shows Bifrost lower at concurrency 1 and 64, Rust lower at 16, a tie at 4, and throughput gaps under 1%. The headline number comes from a separate file with no stated concurrency.
Logging on vs logging off: the configuration gap in LiteLLM's benchmark
LiteLLM's benchmark blog says no gateway logged or persisted anything. The Bifrost v1.6.4 config it published never sets enable_logging or a log store, and the defaults then switch on logging, a SQLite store and built-in plugins. That is inferred from v2.4.0 code, not confirmed on v1.6.4.
A Rust beta vs an old Bifrost build: the version problem in software benchmarking
LiteLLM's Rust gateway benchmark compared an early Rust beta, which its own blog says is still landing streaming, against Bifrost v1.6.4, an old build. The homepage never names the version. Why version parity decides what a benchmark can claim, what the record shows, and a checklist for readers.
Agent benchmarks are now 44% of new AI benchmarks, and most are too small to rank close models
An analysis of 352 sized benchmarks in The Frontier Wire's AI benchmarks dataset finds agent benchmarks are about 4.5 times smaller than the rest. That makes many leaderboard gaps statistical noise. The data, the method, and a table of the smallest score gap worth taking seriously at each benchmark size.
AI energy per prompt estimates span 2,300-fold, but typical text prompts agree at 0.24 to 0.42 Wh
The Frontier Wire compared every per-prompt energy and water figure in its dataset. The extremes differ 2,300-fold for energy and over 1,600-fold for water, but four independent estimates for an ordinary text prompt sit within a factor of 1.75. Scope, request length and modality explain the rest.
AI water use per kWh fell at six big tech firms, yet total consumption rose at all six
An analysis of company sustainability reports in The Frontier Wire's AI water and energy dataset. Every one of the six companies that publish two years of water and electricity used less water per kilowatt-hour. None used less water. Together they added about 17.7 million cubic meters in a year.
Agent skills marketplace options for teams: author, version, review and roll back
A SKILL.md file changes what every coding agent on a team does the next time it loads. How to write descriptions that trigger, keep skills small, version and review them, pick a distribution model from git to plugin marketplaces to a private registry, and roll a bad release back.
AI agent security: a practical guide to securing LLM apps and agents
The model is not a security boundary. Eight concrete failures in LLM apps and agents, from provider keys in browser bundles to injected tool calls and runaway spend, each mapped to the OWASP LLM Top 10, with the code-level fix and an open-source agent skill that audits a codebase for them.
Best Claude Code skills for teams shipping LLM apps and agents
A Claude Code skill is a folder of instructions the agent loads only when a task calls for it. Six MIT-licensed skills for security, observability, cost, MCP, coding-agent governance and skill management, ranked on stated criteria, installed and linted in a clean repository.
Bifrost adds an Agent Gateway that proxies A2A agents beside model and MCP traffic
The open-source gateway's v2.3.0 release, out October 13, registers A2A agents by their agent cards, rewrites those cards to gateway endpoints, bridges transports, relays push notifications and gates every call on virtual-key grants. A v2.3.1 follow-up tightened destination and credential rules.
Bifrost CLI v1.0 launches coding agents through the gateway and manages it from the terminal
Maxim AI has published Bifrost CLI v1.0, a launcher that starts four terminal coding agents already routed through its open-source gateway, scoped to one process. It also signs developers in, shows their models and budget, and exposes the gateway's admin API to scripts and CI.
Bifrost Edge brings AI gateway governance to every company laptop
Maxim AI's endpoint agent for the Bifrost gateway gained per-app capture controls, version reporting and a stateless MCP inspection path in Bifrost Enterprise v2.3.0. How Edge works, what it governs, how it rolls out through MDM, and where it sits next to CASB and DLP tooling.
Bifrost reworks /v1/decisions into one OpenAI-shaped API for decision models
Decision models from TypeSafe, OpenAI, Perplexity, Cloudflare, Microsoft, Databricks and Ollama return typed answers with probabilities, in two request dialects. Bifrost v2.3.0 gives them one endpoint, stops silently answering decisions with chat models, and prices decisions on their own rates.
Bifrost adds Warp, a read-only agent that answers questions about gateway telemetry
Warp, formerly called Odin, puts a tool-calling agent inside the Bifrost gateway's own dashboard. It reads logs, metrics and governance settings through a fixed set of read-only tools, searches logged conversations by meaning, and sees only the rows the person asking could already see.
Claude Code enterprise rollout: managed settings, per-developer keys, budgets and logs
Installing Claude Code takes a minute. Rolling it out to two hundred engineers without a surprise bill, a shared API key or an unknown MCP server on every laptop takes a plan: a provider choice, managed settings, one credential per developer, budgets, CI keys and a way to tell whether it is paying off.
Codex skills: where OpenAI's agent finds SKILL.md files, and which ones to install
A Codex skill is a folder with a SKILL.md file that the CLI and IDE extension find in .agents/skills and load when a task matches its description. Scan paths, the context budget, writing one that triggers, installing shared skills with npx skills, and six worth adding for LLM app teams.
Cursor skills: how Agent Skills work in Cursor, and when to use them instead of rules
Cursor loads Agent Skills from .agents/skills and .cursor/skills and lets the agent decide when each one applies. How that decision works, how skills compare with .cursor/rules and AGENTS.md, how to install shared skills with npx skills, and three skills that fix MCP sprawl and blind spots.
LLM cost optimization: a measured playbook for cutting API spend
Every LLM bill is requests times tokens times price. A playbook for shrinking each factor without a quality regression: attribution, right-sized models, complexity routing, prompt trimming, prompt and response caching, Batch APIs, loop caps and budgets, plus the open-source llm-cost-reduction agent skill.
LLM observability: what to capture, trace and alert on in production
A practical explainer on LLM observability: the per-call record, time to first token, OpenTelemetry GenAI conventions, trace structure for agents, sampling and PII redaction, Prometheus alert thresholds, feedback joins and evals on production traffic, plus the open-source llm-observability-stack agent skill.
MCP best practices: tool budgets, least privilege and approval gates
Connecting one MCP server takes a line of config. Running twenty safely across a team takes tool budgets, per-identity allow-lists, approval gates, sane transports, health checks and logs. The practices, checked against the 2026-07-28 MCP specification, and an open-source agent skill that audits a setup against them.
npx skills and skills.sh: a complete guide to the skills CLI
The skills CLI installs SKILL.md folders from GitHub into 79 coding agents, and skills.sh is the directory built on its telemetry. Every command in version 1.7.4, project and global scope, the symlink layout, the lock file and its hashes, and the warning printed on every install, checked against the source.
API management platforms compared: twelve options for 2026
Every major API management platform now ships some form of AI gateway. The criteria below separate the platforms built to publish and monetize APIs from the layer built to meter tokens, fail over between model providers and govern MCP tools, and compare twelve options against both.
Attribute-based access control (ABAC): a practical guide
NIST defines attribute-based access control as decisions made from attributes of the subject, the object and the environment. The architecture behind it, how it differs from RBAC and Google's Zanzibar-style ReBAC, the policy languages that implement it, and how to apply it to models, agent tools, MCP servers and RAG.
Claude usage: plan limits, how to check them and API tracking
Claude usage is measured three different ways depending on how a request signs in. Subscriptions run on five-hour and weekly windows, API keys run on rate limits and spend caps, and usage-based Enterprise seats are billed per token. Where each meter lives and what drains it, per Anthropic's pages on 21 October 2026.
How to choose an API management solution: a weighted framework
API management products share a feature list and differ in where their planes run, how their policies port and what they can meter. A weighted criteria table, a proof-of-concept plan and a cost model, grounded in the OpenAPI and OAuth specs and OWASP's 2023 API risks, plus the new question of model traffic.
The Model Context Protocol, explained from the specification
MCP is now stateless. Revision 2026-07-28 removed the initialize handshake and protocol sessions, deprecated sampling and roots, and moved server-to-client requests into retried calls. A section-by-section guide to the current specification, its security rules, its registry and SDKs, and how teams run it in production.
OpenRouter models compared: 19 models on price, context and routing
OpenRouter's catalog holds 468 entries under 57 publisher prefixes, but the price next to each name is the price of one endpoint, not of the model. Nineteen leading models compared on published criteria, with picks by job and the routing settings that decide which host, precision and price tier serve a request.
Prompt caching in practice: breakpoints, TTLs and cost math
Every major model API now discounts a repeated prompt prefix, but the rules differ on markers, minimum lengths, lifetimes and write surcharges. A provider-by-provider implementation guide with current prices, the usage fields that prove a hit, worked cost examples and the gateway settings that keep a cache warm.
Rate limiting: a practical guide to algorithms, 429s and LLM limits
Six algorithms, four places to enforce them, one status code and a draft standard for telling clients how much quota is left. A working reference for rate limiting HTTP APIs, with the request and token limits that OpenAI, Anthropic and Google apply to model traffic.
Shadow AI monitoring: a practical guide to detection, inventory and policy
Shadow AI detection is a data problem before it is a blocking problem. The signal sources that reveal unapproved AI tools, what to log and how long to keep it, how to build an AI inventory that NIST and ISO 42001 would recognise, and how to roll monitoring out through MDM without breaking employee privacy law.
User-managed access and rate limiting: a technical guide
User-Managed Access lets a resource owner decide who else may reach their data, through permission tickets and requesting party tokens. Every identity that grant creates is also a quota key. The protocol, the HTTP signals for throttling, and how both apply to LLM APIs and agents acting for users.
A2A vs MCP: how the two agent protocols divide the work
Both protocols now sit under the Linux Foundation's Agentic AI Foundation, and both shipped major revisions in 2026. Read side by side, the specifications draw a clean line: MCP gives one agent depth, A2A gives a system of agents reach.
The 10 best agent observability tools in 2026, compared
An AI agent makes dozens of model and tool calls for one task, and a failure can sit in any of them. Ten observability tools, from gateways that see every call to SDK-based tracing platforms, compared on published criteria and the OpenTelemetry conventions they follow.
The 9 best AI guardrails tools in 2026, compared by where they enforce
AI guardrails now come as gateways, cloud APIs, Python frameworks and small safety models. They differ less in what they detect than in where they sit and what traffic they can see. Nine tools, compared against published criteria, with the gaps each one leaves.
Claude Code pricing: plans, API rates and what drives the bill
Claude Code is included in every paid Claude plan and can also run on pay-as-you-go API keys, cloud accounts or a gateway. Which credential a developer uses decides which bill a request lands on, and context, caching and model choice decide how large that bill gets.
Claude Code security: permissions, sandboxing and the scanner
The phrase now means two things: Anthropic's Claude Security scanner, and the job of keeping a coding agent with shell access from doing harm. Permission modes, the sandbox, managed settings, MCP policy and gateway routing, mapped to the incidents that shaped them.
Claude Code router setups: running Claude Code on other models
Claude Code talks to one API format, and anything that speaks that format can answer it. Routers, gateways and local servers use that to swap in other models. The configuration takes minutes; knowing what breaks on the way takes longer.
Codex vs Claude Code: how the two coding agents compare in 2026
OpenAI's Codex and Anthropic's Claude Code now cover the same ground, from terminal to IDE to cloud sessions. They still differ on defaults that matter: who sandboxes what, how plans meter usage, what is open source and how an administrator locks a fleet down.
LLM API pricing compared: OpenAI, Claude, Gemini, DeepSeek and Grok
Frontier models now cluster around $2 per million input tokens and $10 per million output tokens, but batch discounts, cache multipliers, long-context tiers and off-peak clocks move the real bill by an order of magnitude. Every price below comes from the providers' own pricing pages as of October 15, 2026.
OpenRouter alternatives in 2026: seven gateways compared
OpenRouter made one API key for hundreds of models the default way to prototype. Credit fees, a third party in the data path and governance that lives in someone else's account send production teams looking for alternatives. Seven options, ranked against published criteria.
OWASP LLM Top 10 for 2026: the ten risks and where to enforce each
The 2026 edition of the OWASP Top 10 for LLM applications is the first built partly on incident data. It reorders the list, renames one entry and hands agent risks to a separate Agentic Top 10. Each of the ten risks, with documented examples and the layer that can actually enforce its controls.
Prompt caching vs semantic caching: what each one saves
Provider prompt caches cut the price of a repeated prefix by 90 percent or more but leave output tokens untouched. Semantic caches remove whole calls but can return the wrong answer. The mechanics, prices and failure modes of both, with a worked cost example.
Prompt injection: how the attacks work and which defenses hold
Every major lab, OWASP and the UK's National Cyber Security Centre now say the same thing about prompt injection. Models cannot reliably tell instructions from data, filters miss adaptive attacks, and the defenses that work limit what a manipulated model is able to do.
Self-hosted LLMs: how to run a local LLM, from laptop to GPU cluster
Running a model on your own hardware is now a configuration problem more than a research one. The hard parts are sizing GPU memory, choosing between Ollama, llama.cpp, LM Studio, vLLM and SGLang, reading the licence, and putting a production API in front of a server that was built for one user.
Shadow AI: what it is, how widespread it is, and how to detect it
Two thirds of AI users reach AI services through personal accounts on corporate devices, and shadow AI now appears in 43% of the AI security incidents IBM studied. The risks, the four detection layers, the acceptable use policy and the controls that turn unsanctioned use into governed use.
Tokenmaxxing: why token leaderboards measured the wrong thing
In the spring of 2026, engineers at large technology companies competed on leaderboards ranked by how many AI tokens they consumed. By summer the leaderboards were gone and the budgets were blown. The episode is a lesson in what a token is: a unit of cost, not a unit of work.
Top 5 AI gateways for automatic failover across Bedrock, Vertex AI and Azure OpenAI
Claude now runs on Amazon Bedrock, Google Cloud and Microsoft Foundry, and GPT models run on both OpenAI and Azure. That makes cross-cloud failover practical, but each cloud throttles, authenticates and names models differently. Five gateways, measured against seven published criteria.
Top 5 AI gateways for Claude Code, Codex CLI and Cursor in 2026
Claude Code, Codex CLI and Cursor each reach a model provider in a different way. Five gateways that put all three behind one set of developer keys, budgets and logs, compared on how each client connects, what governance applies and where the gateway runs.
Top 5 AI gateways for multimodal LLM workloads in 2026
A multimodal LLM stack sends images, audio, video and files as well as text. Five gateways, measured endpoint by endpoint against the same published criteria, and which kind of team each one suits.
Top 5 LLM gateways for Anthropic, OpenAI and Gemini in 2026
Most production teams now call at least two of the three big model providers. Five gateways that put OpenAI, Anthropic and Gemini behind one endpoint, measured on native SDK formats, cross-provider failover, spend control and where the proxy runs.
Top 5 AI gateways for routing between vLLM, Ollama and cloud LLMs in 2026
Most teams now run some models on their own GPUs and rent the rest. Five self-hostable gateways that route between vLLM, Ollama and hosted APIs, compared on how they reach local servers, how they decide where a request goes, and what happens when a local box falls over.
Top 5 AI gateways for air-gapped and in-VPC LLM deployments in 2026
An on-premise LLM stack is only as private as the gateway in front of it. Five self-hosted AI gateways, read at the level of what each one fetches from the internet, how it installs without a connection, and how it routes to vLLM, Ollama and private cloud endpoints.
Best agent orchestration frameworks in 2026: 10 open-source options
Agent frameworks have converged on the same feature list, but they still differ in how they model control flow, where state lives and how much they assume about your model provider. Ten code-first, open-source frameworks, compared against published criteria with GitHub activity checked on 5 October 2026.
The 10 best agent orchestration platforms for enterprise in 2026
An enterprise AI agent platform is where agents run, get an identity, reach tools and get audited. Ten managed options, from the three hyperscalers to the SaaS suites and an automation tool, compared on published criteria with documented prices.
Top 10 AI gateways in 2026, ranked on production criteria
An AI gateway now sits in front of most production model traffic. Ten options, from open-source binaries to cloud API platforms and managed routers, measured against ten published criteria before any ranking.
Top 10 AI models in October 2026, ranked by what they do best
Claude Opus 5.5 leads the independent indexes this month, but no single model is best at everything. Ten models, from $50-per-million-token flagships to MIT-licensed open weights, measured against published criteria, with the best pick for reasoning, coding, cost, long context, multimodal and self-hosting.
Top 10 MCP servers in 2026, compared on auth, tools and security
Most of the MCP servers worth connecting in 2026 are now hosted by the vendors themselves, behind OAuth. The differences that matter are how much a server lets an agent change, how finely access can be narrowed, and how each one has handled its security incidents.
The best MCP servers for coding agents in 2026, ranked by workflow
A coding agent is only as useful as the systems it can reach. Ten MCP servers for repositories, browsers, errors, databases, tickets, designs and infrastructure, compared on what they unlock, what they cost in context, and what they can break.
Top 10 open-source AI gateways for self-hosting in 2026
An open-source AI gateway is only as open as its license file and its paid tier allow. Ten self-hosted LLM gateways, read at the level of the LICENSE file, the dependency list and the release history, with criteria published before the ranking.
Top 10 open-weight models in October 2026, ranked
Xiaomi, Z.ai, Moonshot, Alibaba and DeepSeek now publish weights within a few points of the closed frontier. Ten models, ranked on Artificial Analysis and Arena scores, with the licence clause, parameter count and GPU budget that decide whether a team can actually run each one.
Top 10 MCP security tools in 2026, from gateways to scanners
No single product secures the Model Context Protocol. Tool poisoning, rug pulls, shadow servers and token misuse each need a different control. Ten tools, compared against the risks the specification and OWASP name, with the gaps each one leaves.
The 10 best MCP gateways in 2026, compared on security and control
The Model Context Protocol leaves authorization optional and tool safety to the implementer. An MCP gateway is where most teams end up enforcing both. Ten gateways, compared on the controls the specification asks for.
How to use Jev: where a System One model fits next to an LLM
TypeSafe's Jev returns typed answers with probabilities instead of text. Its documentation, the first developer reports and early independent papers show where it fits in a pipeline, what the published price works out to, and why its probabilities need checking on your own data.
LiteLLM alternatives in 2026: gateways compared for production
LiteLLM is the default first gateway for many teams. Its production footprint, the March 2026 PyPI compromise and its license split send some of them looking elsewhere. Six alternatives, measured against published criteria, with a migration path.
LLM cost management: budgets, rate limits and virtual keys at the gateway
Output tokens cost five to six times more than input tokens, reasoning tokens bill as output, and shared API keys erase the link between a charge and its owner. This is how teams attribute and cap LLM spend, and what gateway-level governance in the open-source Bifrost gateway does about it.
LLM failover and load balancing: keeping AI apps up when providers fail
Model APIs fail in documented, predictable ways. The status pages, error codes and rate-limit rules published by OpenAI and Anthropic show which failures a retry fixes, which need a different provider, and which no amount of rerouting will solve.
AI gateway vs API gateway: what changes when the traffic is tokens
API gateways were built for short request-response calls priced per request. Model traffic is priced per token, streamed for minutes and shaped differently by every provider. A mechanism-by-mechanism comparison, grounded in AWS, Kong, NGINX, OpenAI and Anthropic documentation, with Bifrost as the worked example.
System One models, explained: typed answers and calibration
TypeSafe's Jev returns typed answers with probabilities instead of text. The idea borrows Kahneman's System 1, a calibration objective called RLCD, and a strong claim about hallucination. The concepts, the evidence behind them, and where each one stops.
System One models vs LLMs: most AI calls should decide, not talk
Routing, gating, judging and choosing an agent's next step are decisions, not writing. Splitting decision from generation is the right architecture, and TypeSafe's Jev is the first product built for it. Trust will come from independently measured calibration, not a promise that it cannot hallucinate.
TypeSafe launches Jev, a model that returns decisions instead of text
TypeSafe AI came out of stealth on September 15 with $40 million in seed funding and Jev, a model that answers typed questions about a block of text with probabilities instead of prose. The pricing is unusual, the speed claims are large, and the first independent tests are mixed.
The reasoning-model era has a cost problem
Models that think before they answer are billed for every token of that thinking, whether or not you see it. The controls have moved from token budgets to effort levels, and the price of a request now depends on how hard the question is.
Test-time compute, explained: the second scaling lever
The idea that a model can get better by thinking longer, rather than by being bigger, reshaped how labs spend their money. The evidence behind it, method by method, and the limits the papers themselves report.
What an AI gateway does, and what it costs to run one
Most teams that use more than one model provider end up putting a proxy in front of all of them. This is what that proxy does, where the main products differ, and a detailed look at one of them, Bifrost, based on its documentation and a local install.
How to compare LLM evaluation frameworks without getting sold to
Every evaluation framework claims to measure quality. The differences show up in who can write the test, whether a trajectory can fail, and what happens to a bad score in production. Here are the criteria, the research that justifies them, and nine current tools read against them.
Stop calling it hallucination
The word implies a glitch in an otherwise reliable mind. The published evidence says something plainer: models are trained and graded to guess, and the remedy depends on saying so.
The state of open-weight licensing, mapped
Most flagship open-weight releases of the past year ship under Apache 2.0 or MIT. The exceptions carry user caps, revenue gates, prohibited-use policies and remote kill switches. This survey reads the licence texts, not the launch posts, and maps what each one actually permits.
RAG vs. long context: choose by failure mode, not by benchmark
Every frontier API now takes a million tokens, and every published long-context study still finds accuracy falling well before the window is full. Retrieval fails differently, earlier and more visibly. The failure you can tolerate should decide, and current pricing tells you what each choice costs.
Reading a model card like an auditor
The model card was proposed as a one-to-two-page disclosure. Current cards range from a 135-page system card to a benchmark table with a licence. Reading one well means checking it against the original template, the EU AI Act and the Stanford transparency index.