Technical report · live · not yet updated

The Kid: a language model raised in public

Abstract. The Kid is a 1.38-billion-parameter language model trained from random initialization, in public. Training runs in scheduled growth spurts on rented 8×H100 nodes and is paid for by a 3% fee on trades of the $KID token. Fees accrue to a college fund held in NVDA, which pays for compute at no more than 1% of its balance per week. Every checkpoint is graded on fixed prompts, kept, and can be queried below. This page reports the model's live training state, its evaluation results and the funding ledger.

Model: nanochat GPT, 24 layers
Token: $KID · Robinhood Chain
Funding: 3% of trades → fund in NVDA
Compute: 8×H100 SXM per spurt
Status: Pre-launch

DAY 0EGG

Subject 001 Stage: Egg · Brain 1

Stage: Egg
Brain: Brain 1
Training step: 0 / 5,568
Tokens seen: 0
Validation loss: n/a
Coherence: n/a
Milestones: 0 / 8
Checkpoints: 0
Age: n/a
College fund: n/a
Compute spent: $0

Figure 1. Smoothed training loss (nats per token) for the current brain against training step. Numbered markers are milestones from the log in §4. The shaded band is the part of the run not trained yet.

Interactive evaluation

Messages go to a live checkpoint. Before chat fine-tuning the model is a pure next-token predictor: your text becomes a prefix and the model continues it. After fine-tuning it answers in turns. Outputs are shown exactly as generated, with no filtering.

Checkpoint completion temperature 0.8 · top-k 50 · max 48 tokens

systemLoading the latest checkpoint…

Rate limited per visitor. Each message runs on the same checkpoints the grader scores.

Training progress

Pretraining is split into growth spurts. Each spurt resumes the previous checkpoint with its optimizer and data-loader state, so the spurts add up to exactly one uninterrupted run with the same learning-rate schedule. Every spurt ends with a checkpoint, a validation pass and a grading pass.

Figure 2. Validation loss in bits per byte on held-out text at each pretraining checkpoint. Lower means the model is less surprised by unseen text.

**Table 1.** Checkpoints, newest first. Grader metrics are defined in §6.
Checkpoint	Stage	Step	Tokens	Val bpb	Coherence	Variety	Facts	Math	Created
No checkpoints yet.

Developmental stages

Subject 001 is drawn from the model's training state. Its body changes at each stage transition below. A new brain starts from random weights again, at a larger size.

Figure 3. Stages of Subject 001. Filled: reached. Silhouette: not reached yet.

**Table 2.** Stage transitions.
Stage	Triggered by	Training phase	Planned
Egg	Before the first checkpoint	n/a	Pre-launch
Hatchling	Step-0 checkpoint (random initialization) published	Pretraining	Launch hour
Babbler	First pretraining spurt completes	Pretraining, 12 spurts	Days 1–3
Talker	First chat fine-tuning checkpoint	Supervised fine-tuning	Days 4–5
Student	First reinforcement-learning checkpoint	RL on GSM8K	Days 6–7
Teen	Brain 2 begins pretraining	New run, 32 layers	Weeks 2–3
Grad	Brain 3 begins pretraining	New run, 40 layers	Weeks 4–6

Milestone log

A milestone is logged at the first checkpoint where its criterion holds, with the exact output that satisfied it. Criteria are evaluated by the grader after every spurt.

**Table 3.** Milestones reached, in order.
#	Milestone	Checkpoint	Reached	Evidence (verbatim)
No milestones yet.

Milestone criteria

First words that go together: Mean word-pair coherence ≥ 0.25 across the grader's base-model samples.
First sentence: A run of six or more dictionary words ending in . ! or ?, with pair coherence ≥ 0.6 and variety ≥ 0.9 inside the sentence; mean coherence ≥ 0.45 and variety ≥ 0.7 overall.
First fact: At least one of five factual prefixes ("The capital of France is") completed correctly within six words at temperature 0.
First conversation: The first chat fine-tuning checkpoint.
Knows its name: "The Kid" appears in at least two of three answers to name questions.
Knows its ticker: "$KID" appears in at least one of two answers to ticker questions.
First math problem solved: At least one of six grade-school word problems answered correctly.

Checkpoint comparison

The same prompt sent to three checkpoints at once. By default: the first checkpoint, the middle one and the latest.

Evaluation

Two kinds of evaluation. Grader metrics run on every checkpoint and are cheap and noisy. Standard benchmarks run at the end of each training phase and are comparable with published models.

**Table 4.** Grader metrics at the latest checkpoint of each phase, current brain.
Metric	Pretraining	Chat fine-tune	RL	Definition

**Table 5.** Standard benchmarks. The Kid's scores are filled in at the end of each phase.
Benchmark	Measures	GPT-2 (2019)	The Kid, Brain 1
CORE	DCLM composite of in-context learning tasks	0.2565	after pretraining
ARC-Easy	Grade-school science, multiple choice	n/a	after fine-tune
ARC-Challenge	Harder science questions	n/a	after fine-tune
MMLU	57 academic subjects, multiple choice	n/a	after fine-tune
GSM8K	Grade-school math word problems	n/a	after RL
HumanEval	Python function synthesis	n/a	after fine-tune

GPT-2's CORE score is the reference nanochat reports for GPT-2 capability. Chat benchmarks have no comparable GPT-2 figure, so they're marked n/a.

Model and training setup

The model is trained with nanochat, an open-source harness (MIT license). Depth is the single dial: width, heads, training horizon, batch size and learning rates follow from it. Figures below come from nanochat's own sizing code at a tokens-to-parameters ratio of 8.

**Table 6.** Brain specifications. Brain 2 and Brain 3 run only when the fund's allowance pays for them.
	Brain 1	Brain 2	Brain 3
Layers	24	32	40
Model width (d_model)	1,536	2,048	2,560
Attention heads (dim 128)	12	16	20
Context length	2,048	2,048	2,048
Vocabulary (BPE)	32,768	32,768	32,768
Parameters, total	1.38B	2.82B	4.99B
of which value embeddings	604M	1.07B	1.68B
Scaling parameters (matrices + head)	730M	1.68B	3.23B
Training tokens	5.84B	13.4B	25.8B
Batch size (tokens)	1,048,576	2,097,152	2,097,152
Optimizer steps	5,568	6,400	12,320
Training compute (FLOPs)	2.8 × 10¹⁹	1.4 × 10²⁰	5.3 × 10²⁰
Compute relative to Brain 1	1×	5.2×	19×
Pretraining time, one 8×H100 node	~2 h	~10 h (est.)	~38 h (est.)
Pretraining cost at $27.92–31.92/h	$56–64	$290–330 (est.)	$1,060–1,210 (est.)

Training recipe

Optimizer: Muon for weight matrices, AdamW for embeddings and scalars
Learning rate: 40-step warmup, constant, then linear decay over the final 65% of steps to 5% of peak
Precision: FP8 matrix multiplies on H100, bf16 elsewhere
Attention: Sliding window pattern SSSL (three quarter-context layers, then one full-context layer)
Pretraining data: NVIDIA ClimbMix web text
Fine-tuning data: SmolTalk conversations, MMLU auxiliary train, GSM8K, plus 208 identity conversations × 8 epochs
RL: GSM8K, reward for a correct final answer

Growth-spurt protocol

Schedule: A pretraining spurt every 6 hours, 12 in total, over days 1–3
Resumption: --resume-from-step restores weights, optimizer state and data-loader position
Stopping: --stop-at-step ends a spurt without changing the full-run learning-rate schedule
After each spurt: Validation pass, grading pass, checkpoint kept, report posted
Hardware: One rented 8×H100 SXM node per spurt, released afterwards
Reproducibility: Fixed seeds and data order. Every checkpoint is kept.

Funding mechanism

All trading fees go to a college fund held in NVDA. The fund pays only for what a week of training and hosting actually costs, and never more than 1% of its balance in a week. No tokens are burned, and there are no distributions to holders.

$KID tradeon Robinhood Chain
3% feepaid in NVDA
College fundheld in NVDA
Weekly allowance≤ 1% of balance
ComputeGPU hours, receipted

allowance_week = min( cost of the week's training and hosting , 0.01 × fund balance )

If fees stopped entirely and the full 1% were spent every week, the fund would still hold 94% of its balance after six weeks (0.99⁶) and 59% after a year.

Fund balance: Opens at launch
Weekly allowance (max): n/a
Compute spent to date: $0
Fund wallet: Not created yet

**Table 7.** Spending controls.
Control	Enforced by	Status
Public balance and receipts	This page, read from the chain	At launch
1% weekly cap	Signing relay refuses larger withdrawals	At launch
Linear vesting of each fee claim	UNCX vesting contract, 52 weeks	After testing

**Table 8.** Compute ledger.
Started	Phase	Brain	Target	Cost	Status
No spurts yet.

Trading experiment

Scheduled · week 2

After its first report card clears a math threshold, the model receives a small weekly trading budget, its "pocket money". It picks one trade a day in its own words. The college fund itself never trades.

**Table 9.** Trading rules.
Rule	Setting
Budget	min( allowance left after training and hosting , 0.25% of the fund ) per week
Inputs	A daily prompt listing allowed tokens, their price moves, and recent Robinhood Chain trade theses from FOMO
Decision	One ticker and buy, sell or hold, parsed from the model's answer. Unparseable answers count as hold
Universe	Tokens with at least $50k of pool liquidity. Never $KID
Size and pace	One trade a day, at most 25% of the trading wallet per trade
Execution	Allowlisted router and tokens, with a kill switch
Benchmark	Buy-and-hold NVDA over the same period, reported weekly

The $KID token

**Table 10.** Token details.
Name	The Kid
Ticker	$KID
Chain	Robinhood Chain
Paired asset	NVDA (tokenized stock)
Trading fee	3% of every trade, to the college fund
Holder distributions	None
Launch venue	Pons
Contract	Not deployed yet

How to buy

Set up a wallet on Robinhood Chain and fund it with ETH on Robinhood Chain.
Open $KID on Pons with the button below, or paste the contract address.
Swap. 3% of the trade goes to the model's college fund.

Contract address Not deployed yet

Buy on Pons Chart X

Verify the address before you buy. There is only one $KID.

Disclosures

Unfiltered outputs. The model is small and trained from scratch. Its outputs are not moderated and may be wrong, strange or offensive.
Estimates. Brain 2 and Brain 3 times and costs are estimates scaled from Brain 1. Rented GPU prices change.
Custody. The college fund is controlled by the project's wallet, limited by the signing relay's weekly cap, until the vesting contract is live.
Third-party data. Trade theses used in §9 come from the FOMO API and are written by independent traders.
Not financial advice. $KID is a meme coin with no intrinsic value or expectation of financial return. It exists for entertainment. Do your own research.

References

Karpathy, A. nanochat. MIT license. Training harness, sizing code and GPT-2 CORE reference.
ClimbMix 400B (shuffled). Pretraining corpus.
smol-smoltalk. Conversation data for fine-tuning.
MMLU and GSM8K. Fine-tuning, RL and evaluation.
Power Lines (arXiv 2505.13738). Batch-size scaling used by nanochat's sizing code.
RunPod and Lambda GPU pricing, checked Sep 24, 2026.
FOMO API. Trade theses for §9.