Skip to content

registry: Claude-family model identities from harness evidence #72

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
datacraftdevelopment wants to merge 3 commits into
base: main
Choose a base branch
from
Open

registry: Claude-family model identities from harness evidence #72

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
972239c
registry: Claude-family identities (Opus 4.8, Sonnet 5, Haiku 4.5, Fa…
datacraftdevelopment Jul 21, 2026
2bfecab
README: credit datacraftdevelopment for the Claude-family registry en…
datacraftdevelopment Jul 21, 2026
e5a5df6
README: fill in PR number for datacraftdevelopment credit (#72)
datacraftdevelopment Jul 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -357,6 +357,7 @@ Every community PR that lands in main is credited here — that's a project rule
- [@davekopecek](https://github.com/davekopecek) (Dave Kopecek) — committed the design-reference fixture so the design-token guard runs on every machine (#30)
- [@snapsynapse](https://github.com/snapsynapse) (Sam Rogers) — graceful shutdown on SIGINT/SIGTERM with worker-tree cleanup and finished state, plus the 14-test end-to-end CLI regression suite (#4)
- [@mlava](https://github.com/mlava) (Mark Lavercombe) — named setup failures across every diagnostic surface (#37) and `run --baseline`, the no-workers check preflight (#38)
- [@datacraftdevelopment](https://github.com/datacraftdevelopment) — added the Claude-family model identity registry entries (Opus 4.8, Sonnet 5, Haiku 4.5, Fable 5), sourced from Claude Code harness evidence (#72)

Contributions are welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) for the philosophy and what gets a PR merged fast. The short version: small and scoped, rebased on current main, every claim backed by an executed test. Authorship is always preserved — where a maintainer pushes a mechanical fix to your branch, you remain the commit author.

Expand Down
47 changes: 47 additions & 0 deletions registry/model-identity.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,3 +98,50 @@ lab = "Meta"
confidence = "verified"
source = "manifest model field; OpenRouter slug meta-llama/llama-3.3-70b-instruct:free"
last_verified = 2026-07-10

[engines.claude]
harness = "Claude Code"
access = "Claude subscription"
default_model_key = "claude-haiku-4-5"

# Claude-family identities. Evidence precedence is harness-reported model first:
# the Claude Code worker stream self-reports the model it ran in its result
# `modelUsage` telemetry (Ringer's reported_model column stays null only because
# its capture regex is Codex-specific). Each source cites that self-report with
# run names + dates, plus Anthropic's own model page for the lab.

[engines.claude.models."claude-opus-4-8"]
display = "Claude Opus 4.8"
lab = "Anthropic"
confidence = "verified"
source = "Claude Code modelUsage self-report: runs claude-family-audition (2026-07-12), hello-world-bakeoff (2026-07-17); Anthropic models overview https://platform.claude.com/docs/en/about-claude/models/overview"
last_verified = 2026-07-21

[engines.claude.models."claude-sonnet-5"]
display = "Claude Sonnet 5"
lab = "Anthropic"
confidence = "verified"
source = "Claude Code modelUsage self-report: runs claude-family-audition (2026-07-12), hello-world-bakeoff (2026-07-17); Anthropic models overview https://platform.claude.com/docs/en/about-claude/models/overview"
last_verified = 2026-07-21

[engines.claude.models."claude-haiku-4-5"]
display = "Claude Haiku 4.5"
lab = "Anthropic"
confidence = "verified"
source = "Claude Code modelUsage self-report: runs claude-family-audition (2026-07-12), hello-world-site (2026-07-12); Anthropic models overview https://platform.claude.com/docs/en/about-claude/models/overview"
last_verified = 2026-07-21

# Fable 5: Anthropic's Claude 5 announcement names Fable 5 / Mythos 5 as a new
# tier above Opus sharing one underlying model; there is no public lineage beyond
# that. Per docs/TAXONOMY.md, registry aliases are allowed only when the actual
# model lineage is not established, and must be marked as aliases in both the
# registry and the displayed name and never presented as a confirmed lineage.
# The lab (Anthropic) is verified from Anthropic's own announcement; the model
# lineage is not, so this is entered as an unverified alias, not asserted lineage.
[engines.claude.models."claude-fable-5"]
display = "Claude Fable 5 (Anthropic, alias)"
lab = "Anthropic"
alias = true
confidence = "unverified"
source = "Anthropic Claude 5 announcement — Fable 5 / Mythos 5, new tier above Opus, one shared underlying model, no public lineage: https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5; harness self-report (Claude Code modelUsage): runs fable-audition (2026-07-12), crm-build (2026-07-16), chronoline-build (2026-07-17)"
last_verified = 2026-07-21