iChancellor’s AI
← All action models
Open model

Chancellor-1 70B Instruct

Our flagship open Large Action Model

70B params128K tokensApache-2.0Launches September 9, 2026

Not released yet — every artifact below ships on launch day. Until then, the Actionfield preview simulates the experience.

About this model

Chancellor-1 is a fully open 70-billion-parameter Large Action Model (LAM) built for scientific work. It is not a language model that happens to call tools — it is trained end-to-end to plan, decide, and execute multi-step actions, with text as one of its instruments rather than its whole world. And unlike closed frontier models, every part of it is public: the weights, the 4.1T-token training data recipe, the training code, and every intermediate checkpoint.

The design goal was never raw benchmark scores — it was accountable capability. Chancellor-1 was trained with a complete data audit, which means every ability it demonstrates can be traced back to the corpus slices that produced it. That audit is what powers ChancellorTrace in the Actionfield: for any response, you can inspect training documents with exact text matches.

The instruct variant went through three alignment stages, each documented publicly: supervised fine-tuning on openly licensed demonstrations, preference tuning against a released reward model, and a final calibration pass that teaches the model to express uncertainty rather than bluff. The result is a model that says "I'm not sure" measurably more often when it is, in fact, not sure.

Because all 47 intermediate checkpoints are published, Chancellor-1 doubles as a scientific instrument: researchers use the checkpoint series to study how reasoning ability emerges during training — work that is simply impossible with closed models.

What it’s good at

  • Graduate-level science questions and multi-step reasoning
  • Tool calling with open, auditable tool schemas
  • Long-document analysis up to 128K tokens
  • Calibrated uncertainty — it knows what it doesn't know
  • Full reproducibility — retrain it yourself from released artifacts

Where people use it

Research assistants that must cite their sourcesScientific literature review at 128K contextStudying capability emergence via checkpointsHigh-stakes drafting where calibration matters
Method

How it was trained — in the open

01

Corpus assembly & audit

4.1T tokens gathered under documented licenses, deduplicated, and indexed token-by-token so every capability can be traced to its sources.

02

Pretraining

Trained across 47 published checkpoints with full loss curves and data-order logs — the entire run is replayable.

03

Supervised fine-tuning

Instruction demonstrations from openly licensed and consented sources only; the SFT set itself is downloadable.

04

Preference tuning

Aligned against a released open reward model, so the values baked in are inspectable, not implied.

05

Calibration pass

A final stage rewards honest uncertainty, cutting confident-but-wrong answers by a third on held-out evals.

Evidence

Benchmarks, with context

Numbers without baselines are marketing. Every score below ships with its comparison point and its caveat.

BenchmarkThis modelReferenceNote
GradSci-QA (graduate science)71.473.1 · closed frontierwithin 2 points, fully auditable
LiveEval (contamination-resistant)64.858.2 · best open peeritems postdate training cutoff
MathBench68.266.9 · best open peerchain-of-thought, no tools
ToolUse-Hard82.179.4 · closed frontieropen tool schemas
Open artifacts

Everything ships at launch

Launches September 9, 2026. Want a note the moment they’re live? Get notified.

Model weights

safetensors · 140 GBLaunches September 9, 2026

Training data recipe

full audit · 4.1T tokensLaunches September 9, 2026

Training code

Apache-2.0Launches September 9, 2026

Evaluation suite

contamination-resistantLaunches September 9, 2026