diff --git a/.agents/skills/local-review-codex/SKILL.md b/.agents/skills/local-review-codex/SKILL.md
index ec1277fe3a..3483453f66 100644
--- a/.agents/skills/local-review-codex/SKILL.md
+++ b/.agents/skills/local-review-codex/SKILL.md
@@ -1,6 +1,6 @@
 ---
 name: local-review-codex
-description: Run the CI Codex PR review locally against this branch's unpushed work (committed + uncommitted) before pushing. Same policy and reasoning effort as the codex-pr-review GitHub action, on a newer model.
+description: Run the CI Codex PR review locally against this branch's unpushed work (committed + uncommitted) before pushing. Same policy, model and reasoning effort as the codex-pr-review GitHub action.
 ---
 
 # Local Codex Review (pre-push)
@@ -11,18 +11,18 @@ before the PR exists. Use this before `git push` on a non-trivial change.
 
 **Correspondence with CI** — identical:
 - Policy: `REVIEW.md` (severity triage, public-surface checklist, AGENTS.md compliance, test coverage).
+- Model: `gpt-6.1-sol`.
 - Reasoning effort: `model_reasoning_effort="xhigh"`.
 - Output: markdown starting with `## Codex Review`, findings tagged P0 / P1 / P2 with file:line.
 
 **Differences from CI** — local-only:
-- Model is `gpt-6-astra`; CI stays on `gpt-5.6-sol`. Not an oversight to reconcile: `gpt-6-astra` is confirmed on the ChatGPT auth `codex login` uses locally, while CI authenticates with `OPENAI_API_KEY` (`codex-pr-review.yml` prefers it over `CODEX_AUTH_JSON`) and that tier is unverified for the model. Move CI once API access is confirmed, or once CI switches to `CODEX_AUTH_JSON`.
 - Scope is the current branch vs `main` at the merge-base, **including uncommitted changes** (CI reviews a pushed PR diff).
 - Sandbox is `read-only` (CI uses `danger-full-access` on an ephemeral runner). Codex reads the diff and files but cannot modify your working tree.
 - Fresh context is inherent: `codex exec` is a separate cold process, so it does not anchor on the current chat session — the same reason `local-review` insists on a subagent.
 
 ## Prerequisites
 
-- `codex` CLI **>= 0.153.4** installed and authed via `codex login` (an `OPENAI_API_KEY` in the environment takes priority and may not reach `gpt-6-astra` — see the model note above). Older CLIs reject the model with "requires a newer version of Codex"; `run.sh` checks the version up front. Upgrade with `npm install --global @openai/codex@0.153.4` (may need `sudo` for a global install). This matches the pin in `.github/workflows/codex-pr-review.yml` — the CLI version is the same on both sides, only the model differs.
+- `codex` CLI **>= 0.159.3** installed and authed via `codex login` or an `OPENAI_API_KEY` in the environment (which takes priority). Older CLIs reject the model with "requires a newer version of Codex"; `run.sh` checks the version up front. Upgrade with `npm install --global @openai/codex@0.159.3` (may need `sudo` for a global install). This matches the pin in `.github/workflows/codex-pr-review.yml` — the CLI version is the same on both sides.
 - `git fetch` the base ref if it's stale, so the merge-base is accurate.
 
 ## Run
diff --git a/.agents/skills/local-review-codex/run.sh b/.agents/skills/local-review-codex/run.sh
index 948d3820eb..e1fef4ce8e 100755
--- a/.agents/skills/local-review-codex/run.sh
+++ b/.agents/skills/local-review-codex/run.sh
@@ -1,17 +1,14 @@
 #!/usr/bin/env bash
 # Local Codex review — mirrors the .github/workflows/codex-pr-review.yml CI job,
 # but scoped to this branch's unpushed work (committed + uncommitted) so you can
-# review before pushing. Same policy (REVIEW.md) and reasoning effort (xhigh) as CI.
-#
-# The model deliberately differs from CI: gpt-6-astra is confirmed available on the
-# ChatGPT auth `codex login` uses here, but CI authenticates with OPENAI_API_KEY and
-# that tier is unverified for it, so codex-pr-review.yml stays on gpt-5.6-sol.
+# review before pushing. Same policy (REVIEW.md), model and reasoning effort (xhigh)
+# as CI.
 #
 # Usage: run.sh [BASE_REF]   (BASE_REF defaults to "main")
 set -euo pipefail
 
-MODEL="gpt-6-astra"
-CODEX_MIN="0.153.4"
+MODEL="gpt-6.1-sol"
+CODEX_MIN="0.159.3"
 
 BASE_REF="${1:-main}"
 REPO_ROOT="$(git rev-parse --show-toplevel)"
diff --git a/.github/workflows/ai-evals-test.yml b/.github/workflows/ai-evals-test.yml
index 573bc77df9..84b488ae46 100644
--- a/.github/workflows/ai-evals-test.yml
+++ b/.github/workflows/ai-evals-test.yml
@@ -147,7 +147,7 @@ jobs:
           mkdir -p results
           # One cheap model per provider (anthropic/openai/googleai/deepseek).
           fail=0
-          for m in haiku 4o gemini-3-flash-preview deepseek-v4-flash; do
+          for m in haiku gpt-6-luna gemini-3.8-flash deepseek-flash; do
             echo "::group::global-test1-script-create ($m)"
             if ! bun run cli -- run global global-test1-script-create \
                   --model "$m" --execution-only --output "$PWD/results/ci-$m.json"; then
diff --git a/.github/workflows/claude-plan.yml b/.github/workflows/claude-plan.yml
index bc81801417..b3798c76f5 100644
--- a/.github/workflows/claude-plan.yml
+++ b/.github/workflows/claude-plan.yml
@@ -49,7 +49,7 @@ jobs:
           allowed_bots: 'windmill-internal-app[bot]'
           trigger_phrase: '/plan'
           claude_args: |
-            --model claude-opus-5
+            --model claude-opus-5-5
             --system-prompt "# Claude Planning Mode
 
             You are operating in PLANNING MODE ONLY. Your role is to create detailed, structured plans without making any code changes.
diff --git a/.github/workflows/claude.yml b/.github/workflows/claude.yml
index 5e9aab00d3..5250c5a04b 100644
--- a/.github/workflows/claude.yml
+++ b/.github/workflows/claude.yml
@@ -93,4 +93,4 @@ jobs:
             }
           claude_args: |
             --allowedTools "Bash,WebFetch,WebSearch"
-            --model claude-opus-5
+            --model claude-opus-5-5
diff --git a/.github/workflows/codex-pr-review.yml b/.github/workflows/codex-pr-review.yml
index 55036776c6..2d70ad7f7a 100644
--- a/.github/workflows/codex-pr-review.yml
+++ b/.github/workflows/codex-pr-review.yml
@@ -222,7 +222,7 @@ jobs:
 
       - name: Install Codex CLI
         if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
-        run: npm install --global @openai/codex@0.153.4
+        run: npm install --global @openai/codex@0.159.3
 
       - name: Configure Codex auth
         if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
@@ -359,7 +359,7 @@ jobs:
           # e.g. a GitHub Action's index.js, which then runs with our credentials.
           codex exec \
             -C "$GITHUB_WORKSPACE" \
-            -m gpt-5.6-sol \
+            -m gpt-6.1-sol \
             -c 'model_reasoning_effort="xhigh"' \
             -s "$SANDBOX_MODE" \
             -o "$RUNNER_TEMP/codex-final-message.md" \
diff --git a/.github/workflows/pr-ready-review.yml b/.github/workflows/pr-ready-review.yml
index 5e5c6a8c61..90714cba53 100644
--- a/.github/workflows/pr-ready-review.yml
+++ b/.github/workflows/pr-ready-review.yml
@@ -208,4 +208,4 @@ jobs:
             ${{ env.REVIEW_PROMPT }}
           claude_args: |
             --allowedTools "mcp__github_inline_comment__create_inline_comment,Bash(gh pr comment:*),Bash(gh pr diff:*),Bash(gh pr view:*)"
-            --model claude-opus-5
+            --model claude-opus-5-5
diff --git a/AGENTS.md b/AGENTS.md
index 54b5953748..8451a0a600 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -49,7 +49,7 @@ Open-source platform for internal tools, workflows, API integrations, background
 - **Backend patterns**: use the `rust-backend` skill when writing Rust code
 - **Frontend patterns**: use the `svelte-frontend` skill when writing Svelte code. Do NOT edit svelte files unless you have read that skill.
 - **Frontend UUIDs**: do not call `crypto.randomUUID()` in frontend code. Import `randomUUID` from `$lib/utils/uuid` instead.
-- **Code review**: review the current PR or branch against the shared review policy in `REVIEW.md` (severity triage, public-surface checklist, AGENTS.md compliance, test-coverage assessment). The skill at `.agents/skills/local-review/SKILL.md` orchestrates it. All three CLIs auto-discover the same SKILL — Claude reads `.claude/skills/` (symlinked to the canonical `.agents/skills/` file), Codex and Pi read `.agents/skills/` directly. Invoke with `/local-review` in Claude Code, `$local-review` (or `/skills` selector) in Codex, or `pi --skill local-review` / `/skill:local-review` in Pi. For a Codex-driven pass that mirrors the `codex-pr-review` GitHub action against your unpushed work (committed + uncommitted) before you push, use `/local-review-codex` (`.agents/skills/local-review-codex/`) — same `REVIEW.md` policy and `xhigh` reasoning, on `gpt-6-astra` rather than the action's `gpt-5.6-sol`; requires the `codex` CLI >= 0.153.4.
+- **Code review**: review the current PR or branch against the shared review policy in `REVIEW.md` (severity triage, public-surface checklist, AGENTS.md compliance, test-coverage assessment). The skill at `.agents/skills/local-review/SKILL.md` orchestrates it. All three CLIs auto-discover the same SKILL — Claude reads `.claude/skills/` (symlinked to the canonical `.agents/skills/` file), Codex and Pi read `.agents/skills/` directly. Invoke with `/local-review` in Claude Code, `$local-review` (or `/skills` selector) in Codex, or `pi --skill local-review` / `/skill:local-review` in Pi. For a Codex-driven pass that mirrors the `codex-pr-review` GitHub action against your unpushed work (committed + uncommitted) before you push, use `/local-review-codex` (`.agents/skills/local-review-codex/`) — same `REVIEW.md` policy, `gpt-6.1-sol` model and `xhigh` reasoning as the action; requires the `codex` CLI >= 0.159.3.
 - **Domain guides**: `.claude/skills/native-trigger/`
 - **Brand/UI guidelines**: `frontend/brand-guidelines.md`
 - **Domain vocabulary**: `CONTEXT.md` — the words this codebase uses for its own concepts (step, step setting, trigger step, …). Name things the way it does.
diff --git a/ai_evals/core/models.test.ts b/ai_evals/core/models.test.ts
index ba53c24592..d51d61d224 100644
--- a/ai_evals/core/models.test.ts
+++ b/ai_evals/core/models.test.ts
@@ -34,6 +34,10 @@ describe("resolveEvalModel", () => {
 
   it("supports DeepSeek aliases for frontend evals", () => {
     expect(resolveEvalModel("flow", "deepseek").frontend).toEqual({
+      provider: "deepseek",
+      model: "deepseek-flash",
+    });
+    expect(resolveEvalModel("flow", "deepseek-v4-flash").frontend).toEqual({
       provider: "deepseek",
       model: "deepseek-v4-flash",
     });
diff --git a/ai_evals/core/models.ts b/ai_evals/core/models.ts
index 1580cf5b0b..06689c36e1 100644
--- a/ai_evals/core/models.ts
+++ b/ai_evals/core/models.ts
@@ -192,12 +192,21 @@ export const EVAL_MODELS: EvalModelSpec[] = [
   {
     id: "deepseek-v4-flash",
     label: "DeepSeek V4 Flash",
-    aliases: ["deepseek", "deepseek-v4", "deepseek-v4-flash"],
+    aliases: ["deepseek-v4", "deepseek-v4-flash"],
     frontend: {
       provider: "deepseek",
       model: "deepseek-v4-flash",
     },
   },
+  {
+    id: "deepseek-flash",
+    label: "DeepSeek V4.1 Flash",
+    aliases: ["deepseek", "deepseek-flash", "deepseek-v4.1-flash"],
+    frontend: {
+      provider: "deepseek",
+      model: "deepseek-flash",
+    },
+  },
   {
     id: "deepseek-v4-pro",
     label: "DeepSeek V4 Pro",