Skip to content

Commit ee9044e

Browse files
author
1bcMax
committed
v0.12.168: propagate openai/gpt-5.5 across routing, picker, README, SKILL
v0.12.167 added the model entry but every other place ClawRouter advertises a flagship still pointed at gpt-5.4. This pass closes the gap. - src/router/config.ts: insert gpt-5.5 immediately before gpt-5.4 in three fallback chains (auto.COMPLEX, premiumTiers.COMPLEX, agenticTiers.COMPLEX); no primary changes — promoting 5.5 to a primary needs benchmark data we don't have yet - src/index.ts: add openai/gpt-5.5 + anthropic/claude-opus-4.5 to the /model picker allowlist; replace deprecated minimax/minimax-m2.5 with minimax/minimax-m2.7 (cleanup of v0.12.167 oversight in this duplicate list) - README.md: add the openai/gpt-5.5 row to the Premium Models pricing table - skills/clawrouter/SKILL.md: add gpt-5.5 + claude-opus-4.5 to the representative-models line Frozen benchmark docs (docs/smart-llm-router-14-dimension-classifier.md, docs/llm-router-benchmark-46-models-sub-1ms-routing.md) intentionally not touched — adding 5.5 to a benchmark table without measured data would falsify the document.
1 parent 42346bc commit ee9044e

11 files changed

Lines changed: 48 additions & 17 deletions

File tree

‎CHANGELOG.md‎

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -4,6 +4,17 @@ All notable changes to ClawRouter.
44

55
---
66

7+
## v0.12.168 — Apr 25, 2026
8+
9+
- **Propagate `openai/gpt-5.5` everywhere it should appear.** v0.12.167 added the model to `BLOCKRUN_MODELS`, the `gpt-5.5` alias, and the install-script `TOP_MODELS` allowlist — but every other place ClawRouter advertises a flagship still pointed at `gpt-5.4`. This release closes the gap so 5.5 is a first-class citizen across routing, the picker, marketing, and the OpenClaw skill page.
10+
- **`src/router/config.ts` — three fallback-chain insertions, no primary changes.** `openai/gpt-5.5` slots in immediately before `openai/gpt-5.4` in `auto.COMPLEX.fallback`, `premiumTiers.COMPLEX.fallback`, and `agenticTiers.COMPLEX.fallback`. Both stay reachable; 5.5 gets preference when the chain reaches OpenAI. Comments updated so 5.5 is "newest flagship — 1M+ ctx, native agent + computer use" and 5.4 is "previous flagship — benchmarked at 6,213ms, IQ 57". Tier primaries are unchanged: promoting 5.5 to a primary slot needs measured latency/IQ data, which we don't have yet — that's a separate decision tracked outside this release.
11+
- **`src/index.ts` — `/model` picker allowlist updated.** `src/index.ts` carries its own copy of `TOP_MODELS` (separate from the install scripts' identical-but-distinct list — both populate the OpenClaw allowlist depending on install path). Added `openai/gpt-5.5` and `anthropic/claude-opus-4.5` (also missed in v0.12.167's `BLOCKRUN_MODELS` add for opus-4.5), and replaced the now-deprecated `minimax/minimax-m2.5` with `minimax/minimax-m2.7` so the picker matches the deprecation we landed yesterday.
12+
- **`README.md` — Premium Models pricing table.** Added the `openai/gpt-5.5` row at $5.00/$30.00 per 1M tokens (~$0.0175 per 0.5K-in-0.5K-out request), 1M context, full feature set. Placed between `claude-opus-4.6` ($0.0150) and `o1` ($0.0375) so the table stays sorted by approximate $/request.
13+
- **`skills/clawrouter/SKILL.md` — model list line.** The "55+ models including..." line now leads `gpt-5.5, gpt-5.4, ...` and includes `claude-opus-4.5` alongside 4.7/4.6.
14+
- **Files deliberately not touched:** `docs/smart-llm-router-14-dimension-classifier.md` and `docs/llm-router-benchmark-46-models-sub-1ms-routing.md` are frozen benchmark archives — adding 5.5 to a benchmark table without measured numbers would falsify the document. The `posts/*.md` marketing content is similarly point-in-time. Those will be refreshed if/when 5.5 gets benchmarked.
15+
16+
---
17+
718
## v0.12.167 — Apr 24, 2026
819

920
- **Realign the model registry to BlockRun source-of-truth.** Audit found three drifts where ClawRouter's `BLOCKRUN_MODELS` table didn't match what `blockrun/src/lib/models.ts` actually exposes. The server is the source of truth for which models exist and what they cost; the proxy's local view should mirror that 1:1 so cost estimation, the `/model` picker, and routing tier selection all see the same world the server does.

‎README.md‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -340,6 +340,7 @@ Edit existing images with `/img2img`:
340340
| anthropic/claude-sonnet-4.6 | $3.00 | $15.00 | $0.0090 | 200K | reasoning, vision, agentic, tools |
341341
| xai/grok-3 | $3.00 | $15.00 | $0.0090 | 131K | reasoning, tools |
342342
| anthropic/claude-opus-4.6 | $5.00 | $25.00 | $0.0150 | 200K | reasoning, vision, agentic, tools |
343+
| openai/gpt-5.5 | $5.00 | $30.00 | $0.0175 | 1M | reasoning, vision, agentic, tools |
343344
| openai/o1 | $15.00 | $60.00 | $0.0375 | 200K | reasoning, tools |
344345
| openai/gpt-5.2-pro | $21.00 | $168.00 | $0.0945 | 400K | reasoning, tools |
345346
| openai/gpt-5.4-pro | $30.00 | $180.00 | $0.1050 | 400K | reasoning, tools |

‎dist/cli.js‎

Lines changed: 9 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -73437,8 +73437,10 @@ var DEFAULT_ROUTING_CONFIG = {
7343773437
// 1,431ms, IQ 32
7343873438
"google/gemini-2.5-flash",
7343973439
// 1,238ms, IQ 20 — cheap last resort
73440+
"openai/gpt-5.5",
73441+
// Newest OpenAI flagship — 1M+ ctx, native agent + computer use; benchmark TBD
7344073442
"openai/gpt-5.4"
73441-
// 6,213ms, IQ 57 — slowest but highest quality
73443+
// 6,213ms, IQ 57 — previous flagship, benchmarked
7344273444
]
7344373445
},
7344473446
REASONING: {
@@ -73548,8 +73550,10 @@ var DEFAULT_ROUTING_CONFIG = {
7354873550
"moonshot/kimi-k2.6",
7354973551
// Moonshot flagship, independent infra
7355073552
"moonshot/kimi-k2.5",
73553+
"openai/gpt-5.5",
73554+
// Newest OpenAI flagship — 1M+ ctx, native agent + computer use
7355173555
"openai/gpt-5.4",
73552-
// Newest OpenAI flagship (slow but stable)
73556+
// Previous flagship (slow but stable, benchmarked at 6,213ms)
7355373557
"openai/gpt-5.3-codex",
7355473558
"deepseek/deepseek-chat",
7355573559
// Cheap, reliable
@@ -73617,8 +73621,10 @@ var DEFAULT_ROUTING_CONFIG = {
7361773621
// 1,348ms — strong tool use, independent infra
7361873622
"moonshot/kimi-k2.5",
7361973623
// strong tool use, independent infra
73624+
"openai/gpt-5.5",
73625+
// Newest flagship — native agent + computer use (exactly the agentic-tier use case)
7362073626
"openai/gpt-5.4",
73621-
// 6,213ms — slow but reliable flagship
73627+
// Previous flagship — 6,213ms, reliable
7362273628
"deepseek/deepseek-chat",
7362373629
// 1,431ms — cheap, reliable
7362473630
"free/qwen3-coder-480b"

‎dist/cli.js.map‎

Lines changed: 1 addition & 1 deletion
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎dist/index.js‎

Lines changed: 12 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -74045,8 +74045,10 @@ var DEFAULT_ROUTING_CONFIG = {
7404574045
// 1,431ms, IQ 32
7404674046
"google/gemini-2.5-flash",
7404774047
// 1,238ms, IQ 20 — cheap last resort
74048+
"openai/gpt-5.5",
74049+
// Newest OpenAI flagship — 1M+ ctx, native agent + computer use; benchmark TBD
7404874050
"openai/gpt-5.4"
74049-
// 6,213ms, IQ 57 — slowest but highest quality
74051+
// 6,213ms, IQ 57 — previous flagship, benchmarked
7405074052
]
7405174053
},
7405274054
REASONING: {
@@ -74156,8 +74158,10 @@ var DEFAULT_ROUTING_CONFIG = {
7415674158
"moonshot/kimi-k2.6",
7415774159
// Moonshot flagship, independent infra
7415874160
"moonshot/kimi-k2.5",
74161+
"openai/gpt-5.5",
74162+
// Newest OpenAI flagship — 1M+ ctx, native agent + computer use
7415974163
"openai/gpt-5.4",
74160-
// Newest OpenAI flagship (slow but stable)
74164+
// Previous flagship (slow but stable, benchmarked at 6,213ms)
7416174165
"openai/gpt-5.3-codex",
7416274166
"deepseek/deepseek-chat",
7416374167
// Cheap, reliable
@@ -74225,8 +74229,10 @@ var DEFAULT_ROUTING_CONFIG = {
7422574229
// 1,348ms — strong tool use, independent infra
7422674230
"moonshot/kimi-k2.5",
7422774231
// strong tool use, independent infra
74232+
"openai/gpt-5.5",
74233+
// Newest flagship — native agent + computer use (exactly the agentic-tier use case)
7422874234
"openai/gpt-5.4",
74229-
// 6,213ms — slow but reliable flagship
74235+
// Previous flagship — 6,213ms, reliable
7423074236
"deepseek/deepseek-chat",
7423174237
// 1,431ms — cheap, reliable
7423274238
"free/qwen3-coder-480b"
@@ -81848,7 +81854,9 @@ function injectModelsConfig(logger) {
8184881854
"anthropic/claude-sonnet-4.6",
8184981855
"anthropic/claude-opus-4.7",
8185081856
"anthropic/claude-opus-4.6",
81857+
"anthropic/claude-opus-4.5",
8185181858
"anthropic/claude-haiku-4.5",
81859+
"openai/gpt-5.5",
8185281860
"openai/gpt-5.4",
8185381861
"openai/gpt-5.3",
8185481862
"openai/gpt-5.3-codex",
@@ -81860,7 +81868,7 @@ function injectModelsConfig(logger) {
8186081868
"moonshot/kimi-k2.6",
8186181869
"moonshot/kimi-k2.5",
8186281870
"xai/grok-3",
81863-
"minimax/minimax-m2.5",
81871+
"minimax/minimax-m2.7",
8186481872
// Free models (free/ prefix so users see "free" in picker)
8186581873
"free/gpt-oss-120b",
8186681874
"free/gpt-oss-20b",

‎dist/index.js.map‎

Lines changed: 1 addition & 1 deletion
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎package-lock.json‎

Lines changed: 2 additions & 2 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

‎package.json‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
{
22
"name": "@blockrun/clawrouter",
3-
"version": "0.12.167",
3+
"version": "0.12.168",
44
"description": "Smart LLM router — save 85% on inference costs. 55+ models (11 free), one wallet, x402 micropayments.",
55
"type": "module",
66
"main": "dist/index.js",

‎skills/clawrouter/SKILL.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -108,7 +108,7 @@ Rules handle ~80% of requests in <1ms. Only ambiguous queries hit the LLM classi
108108

109109
## Available Models
110110

111-
55+ models including: gpt-5.4, gpt-4o, o3, claude-opus-4.7, claude-opus-4.6, claude-sonnet-4.6, gemini-3.1-pro, gemini-2.5-flash, deepseek-chat, grok-3, kimi-k2.6, kimi-k2.5, and 8 free NVIDIA models (gpt-oss-120b, gpt-oss-20b, deepseek-v3.2, qwen3-coder-480b, glm-4.7, llama-4-maverick, qwen3-next-80b-a3b-thinking, mistral-small-4-119b).
111+
55+ models including: gpt-5.5, gpt-5.4, gpt-4o, o3, claude-opus-4.7, claude-opus-4.6, claude-opus-4.5, claude-sonnet-4.6, gemini-3.1-pro, gemini-2.5-flash, deepseek-chat, grok-3, kimi-k2.6, kimi-k2.5, and 8 free NVIDIA models (gpt-oss-120b, gpt-oss-20b, deepseek-v3.2, qwen3-coder-480b, glm-4.7, llama-4-maverick, qwen3-next-80b-a3b-thinking, mistral-small-4-119b).
112112

113113
## Built-in Agent Tools
114114

‎src/index.ts‎

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -378,7 +378,9 @@ function injectModelsConfig(logger: { info: (msg: string) => void }): void {
378378
"anthropic/claude-sonnet-4.6",
379379
"anthropic/claude-opus-4.7",
380380
"anthropic/claude-opus-4.6",
381+
"anthropic/claude-opus-4.5",
381382
"anthropic/claude-haiku-4.5",
383+
"openai/gpt-5.5",
382384
"openai/gpt-5.4",
383385
"openai/gpt-5.3",
384386
"openai/gpt-5.3-codex",
@@ -390,7 +392,7 @@ function injectModelsConfig(logger: { info: (msg: string) => void }): void {
390392
"moonshot/kimi-k2.6",
391393
"moonshot/kimi-k2.5",
392394
"xai/grok-3",
393-
"minimax/minimax-m2.5",
395+
"minimax/minimax-m2.7",
394396
// Free models (free/ prefix so users see "free" in picker)
395397
"free/gpt-oss-120b",
396398
"free/gpt-oss-20b",

0 commit comments

Comments
 (0)