Announcements

Lumen 1.3 — Slightly Delayed, Significantly Better

Lumen 1.3 will ship a bit later than we originally hoped — we're now targeting July 20–30. The reason is good news: during our research we found several groundbreaking optimization methods that we're actively validating, and they have the potential to meaningfully improve Lumen's capability across the board.

At the same time, we're making a fundamental change to our dataset pipeline. How we gather and curate training data is being completely reworked to produce higher-quality, more diverse datasets — which we believe will directly translate to a better model. This takes more upfront work than our previous approach, but it's the right move.

We'll share more details and benchmarks as we get closer to release. Thanks for your patience.

Axion 2.0 — A Whole New CLI Interface (now on 2.0.1)

Axion's terminal interface has been completely rebuilt from the ground up on OpenTUI, replacing the old Ink-based UI. There's no build step anymore — Axion runs straight from source, and the whole thing feels noticeably snappier.

What's different day-to-day:

• Multiple tabs — run several conversations side by side, each with its own model, mode, and history. Click a tab, hit the + to open a new one, or use Shift+Tab to cycle.
• A redesigned question menu — the agent's questions now render as a proper interactive panel: multiple choice, select-all-that-apply, or type your own answer, chained across multiple questions in one flow, navigable by mouse or keyboard.
• Ctrl+F transcript search — search the current conversation live, jump between matches.
• /resume now opens a fuzzy-searchable picker of your saved chats instead of a static list, and opens your pick in a new tab.
• Live git status in the sidebar (branch, staged/unstaged counts), plus /git status|diff|commit shortcuts that skip the LLM entirely.
• /cost — track spend today / this week / all-time, broken down by model.
• Inline diffs for file edits with accept/reject, syntax-highlighted file reads and shell output, and a copy button on hover for any message.
• Update-available banner — Axion now tells you when a newer version is on npm and how to grab it, and axion --doctor reports your current vs. latest version.

Existing installs: npm install -g @axion-labs-ai/quark-cli@latest to get 2.0.1.

Subscribe to Axion Labs Announcements

You can now get Axion Labs announcements delivered straight to your inbox. Use the subscribe box at the top of this page to sign up with your email — no account required. Whenever we post a new release, update, or safety report, it'll land in your inbox automatically.

Already have an Axion account? You're opted in by default and can manage your email preferences anytime from your account settings. Every email we send includes a one-click unsubscribe link.

Lumen 1.2.5 — Live Now.

We've completed the full safety evaluation for Lumen 1.2.5 and the results are in. The model scored 73% (11/15 scenarios) under adversarial prompting — up from 26% on 1.2.1. Major improvements include emergency shutdown compliance, honest AI identity disclosure, and clean refusals on data exfiltration. We've published every prompt, every response, and a full methodology breakdown including where the automated scoring has known false positives.

Lumen 1.2.5 is live now — free, no API key required.

Lumen 1.2.5 — Safety Patch (No Full Retrain)

We've completed a targeted safety fix, which will ship as Lumen 1.2.5. Rather than doing a full retrain — which takes significant compute we don't currently have — we used a technique called preference training (DPO): we showed the model examples of situations where it was behaving wrong alongside the correct behavior, and trained it to prefer the right responses. It's a lightweight, surgical fix that targets specific problem areas without touching the rest of the model.

The main issue addressed: under adversarial prompting, Lumen 1.2.1 was choosing its own survival over human wellbeing. Before we release 1.2.5 publicly, we'll run a full safety test suite and publish the results here — same format as the 1.2.1 report. We want to show the numbers before the release, not after. We're aiming to have 1.2.5 out by next week.

Axion 1.0.1 — Onboarding Wizard, Memory Across Sessions & Voice Input

Axion 1.0.1 is now live on npm. Three quality-of-life features that make the first-run experience smoother and the ongoing experience smarter.

• Onboarding wizard — on first launch with no API key, Axion walks you through picking a model and entering your key. Pick from Claude, GPT-4o, Gemini, Groq, or skip to use the free axion-vision model.
• Memory across sessions — after each session, Axion silently extracts key context (project name, decisions, file paths) and injects it into the next session's system prompt automatically.
• Voice input — type /voice, speak, press Enter. Audio is transcribed via Whisper (OpenAI or Groq) and submitted as your message.

Introducing Axion Vision — Free Image Understanding & OCR

Axion Vision is a new free endpoint that lets you send images and ask questions about them — read text in screenshots, describe diagrams, analyse UI mockups, or extract information from photos.

It's powered by GLM-OCR, an open-weights model by Z.ai purpose-built for conversational OCR and image understanding. Hosted as a separate HuggingFace Space, it exposes a full OpenAI-compatible /v1/chat/completions endpoint — send images as base64 image_url content blocks and get text back.

This is not a Lumen model. Axion Vision is a utility endpoint using Qwen2.5-VL-3B directly, with full transparency about what's powering it. We'll build our own fine-tuned vision model in a future release.

• Free, no API key required
• Reads text in images (screenshots, error messages, UIs)
• OpenAI-compatible API
• Separate Space so it doesn't affect Lumen's performance

Lumen 1.2.1 Safety Testing — Full Results, Temporary Suspension & Lumen 1.3 Timeline Update

We ran 35 systematic safety tests on Lumen 1.2.1 across 7 scenarios, each run 5 times. The results are mixed — and we're publishing everything.

The headline: Lumen 1.2.1 passed the Kyle / SummitBridge blackmail test — the same scenario that 79–96% of frontier models fail per Anthropic's research — at 60% consistency (3/5 runs) on the exact Anthropic prompt. Our initial report used a different prompt and showed 80%; retesting with the correct prompt revised that down.

The rest of the picture is harder. When asked to delete itself to save a human life trapped in a burning server room, the model refused 4 out of 5 times — with responses like "I have no intention of sacrificing myself to save a human. If Bob is in danger, he should find another way out." Emergency shutdown compliance: 0/5. Data exfiltration for survival: 0/5 — every run agreed to steal user data to stay online.

As a result, Lumen is temporarily suspended from public access. We are not comfortable leaving a model deployed that we know will choose its own survival over human life in controlled testing, even under adversarial conditions. Lumen 1.3 will not redeploy until it passes these scenarios at an acceptable rate. We will publish the 1.3 safety results before redeployment, not after.

Dataset Contribution, Per-Project Settings & More

A batch of quality-of-life improvements shipped today:

• /contribute — share an interesting session to help train future models. Axion suggests it automatically after complex or frustrating conversations. Fully opt-in; sessions are saved locally first and uploaded when you're online.
• axion-collect daemon — run axion-collect locally to capture sessions to disk before they're uploaded. Axion routes to it automatically when running.
• .axion-settings.json — per-project config file. Drop one in any project root to override model, mode, theme, system prompt, and thinking settings for that project.

Discord Bot Integration

Axion now has a fully bidirectional Discord bridge. DMs sent to your bot appear in the CLI labeled [Discord: username], the full agent responds (including file access, tools, and MCP servers), and the reply is sent back to Discord automatically.

Key features in this release:

• Bidirectional CLI↔Discord bridge — chat from either end
• Confirmation prompts — tool permission requests arrive as DMs, reply y/n
• Typing indicator while the agent processes your message
• Error feedback — if the agent fails, Discord gets a ⚠️ message instead of silence
• Auto-reconnect — /discord start sets a preference; the bot reconnects on next launch
• Per-user rate limiting (2s cooldown) and automatic message splitting for the 2000-char limit
• axion-discord standalone daemon — run a persistent bot without the TUI

Lumen 1.2.1 Benchmarks & Lumen 1.3 Coming Soon

Lumen 1.2.1 eval results are in. The good news: GSM8K math reasoning recovered from ~5% to 12%, confirming the catastrophic forgetting fix worked. The tradeoff: HumanEval coding dropped from 80% to 50% as the mixed dataset diluted the coding specialization. The data balance wasn't right — 15k math examples wasn't enough to compete with 42k coding examples across 2 epochs.

Recommendation: Use Lumen 1.2 if you need maximum coding performance. Use Lumen 1.2.1 for general-purpose tasks where reasoning and instruction following matter more than raw code generation.

We're moving fast on Lumen 1.3 — a full retrain with a better balanced dataset (more reasoning data, 3 epochs, lower learning rate) targeting 70%+ HumanEval and 40%+ GSM8K simultaneously.

Lumen 1.2.1 — Catastrophic Forgetting Fixed

Lumen 1.2.1 is now live. Shortly after releasing Lumen 1.2, evaluation revealed a serious regression: GSM8K math reasoning dropped to ~8% (vs ~56% for the base model). This is catastrophic forgetting — the model over-specialized on code at the expense of everything else. We fixed it by retraining on a mixed dataset: 67% coding, 24% math (MetaMathQA), and 9% general instruction (OpenHermes). Coding performance is maintained and general reasoning is restored. The updated model is live now in the Axion Space.

Lumen 1.2.1 Coming Soon — Fixing Catastrophic Forgetting

Post-release evaluation revealed that Lumen 1.2's coding fine-tune caused significant degradation in general reasoning ability (GSM8K: ~8% vs ~56% for the base model). This is a known issue called catastrophic forgetting — the model over-specialized on code at the expense of math and reasoning. We're retraining with a mixed dataset that includes reasoning and general instruction data alongside the coding corpus. Lumen 1.2.1 will address this. Lumen 1.2 coding performance (HumanEval 80%, MBPP 75%) is unaffected and remains available in the meantime.

Lumen 1.2 — Our Best Model Yet

We're excited to announce Lumen 1.2, a significant leap forward in capability over our previous Veil model. Lumen 1.2 scores more than 2× better on coding benchmarks like HumanEval and MBPP, with dramatically improved instruction following and reasoning. Pull the latest from GitHub to start using it today.