Replace retired Claude model ids in rvagent defaults and demo scripts - #997
Open
kasparovabi wants to merge 1 commit into
Open
kasparovabi wants to merge 1 commit into
kasparovabi wants to merge 1 commit into
Conversation
claude-sonnet-4-20250514 and claude-3-5-haiku-20241022 are on Anthropic's retirement list. The CLI and WASM defaults move to claude-sonnet-5, the tester demo to claude-haiku-4-5-20251001. Mock fixtures in anthropic.rs are left as they are since they never reach the network.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
rvagentdefaults toanthropic:claude-sonnet-4-20250514and the three demo scripts undercrates/rvAgent/examples/sendclaude-sonnet-4-20250514orclaude-3-5-haiku-20241022straight toapi.anthropic.com. Both ids are on Anthropic's retirement list (model deprecations), so a freshrvagentrun with no--modelflag, and each of the demos, now get an error back from the API instead of a completion.rvagent-cli/src/main.rs(--modeldefault, and the test that asserts it)anthropic:claude-sonnet-4-20250514anthropic:claude-sonnet-5rvagent-wasm/src/lib.rs(default_model(), its test, two doc comments)anthropic:claude-sonnet-4-20250514anthropic:claude-sonnet-5rvagent-backends/tests/live_anthropic_test.rsanthropic:claude-sonnet-4-20250514anthropic:claude-sonnet-5examples/demo_coder_agent.sh,examples/demo_security_agent.shclaude-sonnet-4-20250514claude-sonnet-5examples/demo_tester_agent.shclaude-3-5-haiku-20241022claude-haiku-4-5-20251001Nothing else changes.
resolve_modelalready splits on the first colon, so the new id needs no code path of its own. The backend omitstemperaturewhen the config value is0.0(the default), which matters because the Sonnet 5 endpoint rejects non-default sampling parameters. None of the demo payloads set one.Left alone on purpose. The mock fixtures inside
rvagent-backends/src/anthropic.rs(test_config()and the serialized request/response samples) still sayclaude-sonnet-4-20250514. They never reach the network, they only check wire format, and I did not want to churn a hundred lines of test JSON for a string that has no effect there. Same forrvagent-cli/tests/integration_tests.rs,session.rsandapp.rs, where the id is a literal passed in by the test rather than a default.Verified locally with
cargo test -p rvagent-cli --bin rvagent test_cli_parse_defaultsandcargo test -p rvagent-wasm --lib test_config_defaults. I have not run the live test or the demo scripts against the API, the claim rests on the published retirement dates.I read this diff line by line before opening it, and I am happy to adjust or drop any part of it. What brought me here was a checker I maintain that flags retired Claude model ids, so I am mentioning that up front rather than leaving you to wonder. If you would rather handle this your own way, or would rather not get this kind of contribution at all, tell me and I will not send another.