@blockrun/clawrouter 0.12.240 → 0.12.241

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (10) hide show
  1. package/README.md +16 -17
  2. package/dist/cli.js +29 -22
  3. package/dist/index.js +29 -22
  4. package/docs/anthropic-cost-savings.md +1 -1
  5. package/docs/anthropic-third-party-harness-changes.md +2 -2
  6. package/docs/clawrouter-cuts-llm-api-costs-500x.md +1 -1
  7. package/docs/clawrouter-vs-openrouter-llm-routing-comparison.md +3 -3
  8. package/package.json +2 -2
  9. package/skills/clawrouter/SKILL.md +3 -3
  10. package/src/top-models.json +0 -1
package/README.md CHANGED
@@ -7,11 +7,11 @@
7
7
  <p>Agents can't sign up for accounts. Agents can't enter credit cards.<br>
8
8
  Agents can only sign transactions.<br><br>
9
9
  <strong>ClawRouter is the only LLM router that lets agents operate independently.</strong><br><br>
10
- <em><!-- br:models.free -->7<!-- /br:models.free --> models free, no crypto required. No signup. No API key. No credit card.</em></p>
10
+ <em><!-- br:models.free -->6<!-- /br:models.free --> models free, no crypto required. No signup. No API key. No credit card.</em></p>
11
11
 
12
12
  <br>
13
13
 
14
- <img src="https://img.shields.io/badge/🆓_7_Free_Models-success?style=for-the-badge" alt="7 free models">&nbsp;
14
+ <img src="https://img.shields.io/badge/🆓_6_Free_Models-success?style=for-the-badge" alt="6 free models">&nbsp;
15
15
  <img src="https://img.shields.io/badge/🤖_Agent--Native-black?style=for-the-badge" alt="Agent native">&nbsp;
16
16
  <img src="https://img.shields.io/badge/🔑_Zero_API_Keys-blue?style=for-the-badge" alt="No API keys">&nbsp;
17
17
  <img src="https://img.shields.io/badge/⚡_Local_Routing-yellow?style=for-the-badge" alt="Local routing">&nbsp;
@@ -34,7 +34,7 @@ Agents can only sign transactions.<br><br>
34
34
 
35
35
  </div>
36
36
 
37
- > **ClawRouter** is an open-source smart LLM router that reduces AI API costs by up to <!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->%. It analyzes each request across <!-- br:clawrouter.dimensions -->15<!-- /br:clawrouter.dimensions --> dimensions and routes to the cheapest capable model in under 1ms, entirely locally. ClawRouter is the only LLM router built for autonomous AI agents — it uses wallet signatures for authentication (no API keys) and USDC micropayments via the x402 protocol (no credit cards). <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models from OpenAI, Anthropic, Google, xAI, DeepSeek, and more. MIT licensed.
37
+ > **ClawRouter** is an open-source smart LLM router that reduces AI API costs by up to <!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->%. It analyzes each request across <!-- br:clawrouter.dimensions -->15<!-- /br:clawrouter.dimensions --> dimensions and routes to the cheapest capable model in under 1ms, entirely locally. ClawRouter is the only LLM router built for autonomous AI agents — it uses wallet signatures for authentication (no API keys) and USDC micropayments via the x402 protocol (no credit cards). <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models from OpenAI, Anthropic, Google, xAI, DeepSeek, and more. MIT licensed.
38
38
 
39
39
  ---
40
40
 
@@ -46,7 +46,7 @@ Every other LLM router was built for **human developers** — create an account,
46
46
 
47
47
  ClawRouter is built for the agent-first world:
48
48
 
49
- - **Starts at $0** — <!-- br:models.free -->7<!-- /br:models.free --> NVIDIA models are free forever (incl. 1M-context DeepSeek V4 Flash + a vision-capable Nemotron Omni)
49
+ - **Starts at $0** — <!-- br:models.free -->6<!-- /br:models.free --> NVIDIA models are free forever (incl. 1M-context DeepSeek V4 Flash + a vision-capable Nemotron Omni)
50
50
  - **No accounts** — a wallet is generated locally, no signup
51
51
  - **No API keys** — your wallet signature IS authentication
52
52
  - **No model selection** — <!-- br:clawrouter.dimensions -->15<!-- /br:clawrouter.dimensions -->-dimension scoring picks the right model automatically
@@ -61,8 +61,8 @@ This is the stack that lets agents operate autonomously: **x402 + USDC + local r
61
61
 
62
62
  | | OpenRouter | LiteLLM | Martian | Portkey | **ClawRouter** |
63
63
  | ---------------- | ----------------- | ---------------- | ----------------- | ----------------- | ---------------------------------------------------------------------- |
64
- | **Models** | 200+ | 100+ | Smart routing | Gateway | **<!-- br:models.chatVisible -->65<!-- /br:models.chatVisible -->** |
65
- | **Free tier** | Rate-limited | BYO keys | No | No | **<!-- br:models.free -->7<!-- /br:models.free --> models, no signup** |
64
+ | **Models** | 200+ | 100+ | Smart routing | Gateway | **<!-- br:models.chatVisible -->71<!-- /br:models.chatVisible -->** |
65
+ | **Free tier** | Rate-limited | BYO keys | No | No | **<!-- br:models.free -->6<!-- /br:models.free --> models, no signup** |
66
66
  | **Routing** | Manual selection | Manual selection | Smart (closed) | Observability | **Smart (open source)** |
67
67
  | **Auth** | Account + API key | Your API keys | Account + API key | Account + API key | **Wallet signature** |
68
68
  | **Payment** | Credit card | BYO keys | Credit card | $49-499/mo | **USDC per-request** |
@@ -78,7 +78,7 @@ This is the stack that lets agents operate autonomously: **x402 + USDC + local r
78
78
 
79
79
  ## Quick Start
80
80
 
81
- > **No wallet? <!-- br:models.free -->7<!-- /br:models.free --> models work free out of the box.** Install, run, and pin `free/mistral-nemotron` (or any of the <!-- br:models.free -->7<!-- /br:models.free -->) — no crypto, no signup, no balance required. Add USDC later when you want paid models.
81
+ > **No wallet? <!-- br:models.free -->6<!-- /br:models.free --> models work free out of the box.** Install, run, and pin `free/mistral-nemotron` (or any of the <!-- br:models.free -->6<!-- /br:models.free -->) — no crypto, no signup, no balance required. Add USDC later when you want paid models.
82
82
 
83
83
  ### Option A — OpenClaw Agent
84
84
 
@@ -107,7 +107,7 @@ openclaw gateway restart
107
107
 
108
108
  > **Using Claude Code?** Check out [BRCC](https://blockrun.ai/brcc.md) — it's purpose-built for Claude Code with the same smart routing and x402 payments.
109
109
  >
110
- > **Using NousResearch Hermes?** See [ClawRouter-Hermes](https://github.com/BlockRunAI/ClawRouter-Hermes) — a Python plugin that wires Hermes into the ClawRouter proxy. Same wallet, same <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, same x402 USDC settlement on Base & Solana.
110
+ > **Using NousResearch Hermes?** See [ClawRouter-Hermes](https://github.com/BlockRunAI/ClawRouter-Hermes) — a Python plugin that wires Hermes into the ClawRouter proxy. Same wallet, same <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, same x402 USDC settlement on Base & Solana.
111
111
 
112
112
  No OpenClaw required. ClawRouter runs as a local proxy on port 8402.
113
113
 
@@ -118,7 +118,7 @@ npx @blockrun/clawrouter
118
118
  ```
119
119
 
120
120
  **2. Fund your wallet** — optional, skip for free tier
121
- Your wallet address is printed on first run. For paid models, send a few USDC on Base or Solana — $5 covers thousands of requests. To stay at $0, pin any of the <!-- br:models.free -->7<!-- /br:models.free --> free models (e.g. `free/mistral-nemotron`) or use `/model free` inside OpenClaw.
121
+ Your wallet address is printed on first run. For paid models, send a few USDC on Base or Solana — $5 covers thousands of requests. To stay at $0, pin any of the <!-- br:models.free -->6<!-- /br:models.free --> free models (e.g. `free/mistral-nemotron`) or use `/model free` inside OpenClaw.
122
122
 
123
123
  **3. Point your client at `http://localhost:8402`**
124
124
 
@@ -391,7 +391,7 @@ No Surf account, no API key — settles directly to Surf's Base treasury in USDC
391
391
 
392
392
  ## Models & Pricing
393
393
 
394
- <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models across 9 providers, one wallet. **<!-- br:models.free -->7<!-- /br:models.free --> models are $0 — paid models start at fractions of a cent.**
394
+ <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models across 9 providers, one wallet. **<!-- br:models.free -->6<!-- /br:models.free --> models are $0 — paid models start at fractions of a cent.**
395
395
 
396
396
  > **💡 "Cost per request"** = estimated cost for a typical chat message (~500 input + 500 output tokens). Paid requests also carry a flat **$0.001/tx settlement fee** (covers on-chain gas; already included in the price the gateway quotes). Free models never pay it.
397
397
 
@@ -400,7 +400,6 @@ No Surf account, no API key — settles directly to Surf's Base treasury in USDC
400
400
  | Model | Input $/M | Output $/M | ~$/request | Context | Features |
401
401
  | ------------------------------------------- | --------: | ---------: | ---------: | ------- | -------------------------------------------- |
402
402
  | free/deepseek-v4-flash | **FREE** | **FREE** | **$0** | 1M | reasoning + coding, huge context |
403
- | free/seed-oss-36b | **FREE** | **FREE** | **$0** | 131K | coding |
404
403
  | free/nemotron-3-nano-omni-30b-a3b-reasoning | **FREE** | **FREE** | **$0** | 256K | reasoning, **vision** (text+img+video+audio) |
405
404
  | free/mistral-nemotron | **FREE** | **FREE** | **$0** | 131K | instruction following |
406
405
  | free/step-3.7-flash | **FREE** | **FREE** | **$0** | 131K | reasoning |
@@ -474,7 +473,7 @@ No Surf account, no API key — settles directly to Surf's Base treasury in USDC
474
473
  | openai/gpt-5.4-pro | $30.00 | $180.00 | $0.1050 | 400K | reasoning, tools |
475
474
  | openai/gpt-5.5-pro | $30.00 | $180.00 | $0.1050 | 1M | reasoning, vision, tools |
476
475
 
477
- > **Free tier:** several NVIDIA-hosted models cost nothing — `/model free` smart-routes across them, or pick one directly (e.g., `/model nemotron-omni` for vision, `/model v4-flash` for 1M-context reasoning, `/model seed-oss` for coding, `/model mistral-large` for the 675B flagship).
476
+ > **Free tier:** several NVIDIA-hosted models cost nothing — `/model free` smart-routes across them, or pick one directly (e.g., `/model nemotron-omni` for vision, `/model v4-flash` for 1M-context reasoning, `/model mistral-large` for the 675B flagship).
478
477
  > **Best value:** `gpt-5-nano` and `gemini-2.5-flash-lite` deliver strong results at ~$0.0003/request.
479
478
 
480
479
  ---
@@ -645,7 +644,7 @@ npm test
645
644
 
646
645
  **The LLM router built for autonomous agents**
647
646
 
648
- You're here. <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, local smart routing, x402 USDC payments — the only stack that lets agents operate independently.
647
+ You're here. <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, local smart routing, x402 USDC payments — the only stack that lets agents operate independently.
649
648
 
650
649
  `curl -fsSL https://blockrun.ai/ClawRouter-update | bash`
651
650
 
@@ -656,7 +655,7 @@ You're here. <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> mod
656
655
 
657
656
  **BlockRun for Claude Code**
658
657
 
659
- Run Claude Code with <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, no rate limits, no Anthropic account, no phone verification. Pay per request with USDC — your wallet is your identity.
658
+ Run Claude Code with <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, no rate limits, no Anthropic account, no phone verification. Pay per request with USDC — your wallet is your identity.
660
659
 
661
660
  `curl -fsSL https://blockrun.ai/brcc-install | bash`
662
661
 
@@ -669,7 +668,7 @@ Run Claude Code with <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible
669
668
 
670
669
  **ClawRouter for NousResearch Hermes**
671
670
 
672
- Python plugin that wraps the ClawRouter proxy for `hermes-agent`. Same <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, same x402 USDC payments on Base & Solana, native Hermes ergonomics.
671
+ Python plugin that wraps the ClawRouter proxy for `hermes-agent`. Same <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, same x402 USDC payments on Base & Solana, native Hermes ergonomics.
673
672
 
674
673
  `pip install hermes-plugin-clawrouter`
675
674
 
@@ -690,7 +689,7 @@ Python plugin that wraps the ClawRouter proxy for `hermes-agent`. Same <!-- br:m
690
689
  | ------------------------------------------------------ | ----------------------------------------------------------------------- |
691
690
  | [Documentation](https://blockrun.ai/docs) | Full docs |
692
691
  | [Model Pricing](https://blockrun.ai/models) | All models & prices |
693
- | [Image Generation & Editing](docs/image-generation.md) | API examples, <!-- br:models.image -->8<!-- /br:models.image --> models |
692
+ | [Image Generation & Editing](docs/image-generation.md) | API examples, <!-- br:models.image -->9<!-- /br:models.image --> models |
694
693
  | [Routing Profiles](docs/routing-profiles.md) | ECO/AUTO/PREMIUM details |
695
694
  | [Architecture](docs/architecture.md) | Technical deep dive |
696
695
  | [Configuration](docs/configuration.md) | Environment variables |
@@ -733,7 +732,7 @@ ClawRouter works with any tool that makes OpenAI-compatible API calls — point
733
732
 
734
733
  ### Is ClawRouter free?
735
734
 
736
- ClawRouter itself is free and MIT licensed. You pay only for the LLM API calls routed through it — and several NVIDIA-hosted models (`deepseek-v4-flash`, `seed-oss-36b`, `nemotron-3-nano-omni-30b-a3b-reasoning`, `mistral-nemotron`, `step-3.7-flash`, `nemotron-nano-9b-v2`, `nemotron-nano-12b-v2-vl`) are completely free. Use `/model free` to smart-route across them, or pick any by name.
735
+ ClawRouter itself is free and MIT licensed. You pay only for the LLM API calls routed through it — and several NVIDIA-hosted models (`deepseek-v4-flash`, `nemotron-3-nano-omni-30b-a3b-reasoning`, `mistral-nemotron`, `step-3.7-flash`, `nemotron-nano-9b-v2`, `nemotron-nano-12b-v2-vl`) are completely free. Use `/model free` to smart-route across them, or pick any by name.
737
736
 
738
737
  ---
739
738
 
package/dist/cli.js CHANGED
@@ -34773,8 +34773,8 @@ var init_config = __esm({
34773
34773
  // FREE — smaller, faster
34774
34774
  "free/deepseek-v4-flash",
34775
34775
  // FREE — 1M ctx; slow (~10 tok/s, 07-28 probe) but completes
34776
- "free/seed-oss-36b",
34777
- // FREE — live coder (qwen3.5-122b + qwen3-next died 2026-07; mistral-large-3-675b EOL'd 07-28)
34776
+ // seed-oss-36b sat here as the free coder until it EOL'd 2026-08-03 (HTTP 410).
34777
+ // gpt-oss-120b/20b already head this chain, so the rung is dropped, not retargeted.
34778
34778
  "google/gemini-3.1-flash-lite",
34779
34779
  // $0.25/$1.50 — newest flash-lite
34780
34780
  "openai/gpt-5.4-nano",
@@ -34891,8 +34891,8 @@ var init_config = __esm({
34891
34891
  "openai/gpt-5.3-codex",
34892
34892
  "deepseek/deepseek-chat",
34893
34893
  // Cheap, reliable
34894
- "free/seed-oss-36b"
34895
- // NVIDIA free ultimate backstop (qwen3-coder-480b retired)
34894
+ "free/gpt-oss-120b"
34895
+ // NVIDIA free ultimate backstop (was seed-oss-36b; EOL'd 2026-08-03)
34896
34896
  ]
34897
34897
  },
34898
34898
  REASONING: {
@@ -34981,8 +34981,8 @@ var init_config = __esm({
34981
34981
  // Previous flagship — 6,213ms, reliable
34982
34982
  "deepseek/deepseek-chat",
34983
34983
  // 1,431ms — cheap, reliable
34984
- "free/seed-oss-36b"
34985
- // NVIDIA free ultimate backstop (qwen3-coder-480b retired)
34984
+ "free/gpt-oss-120b"
34985
+ // NVIDIA free ultimate backstop (was seed-oss-36b; EOL'd 2026-08-03)
34986
34986
  ]
34987
34987
  },
34988
34988
  REASONING: {
@@ -35090,7 +35090,6 @@ var init_top_models = __esm({
35090
35090
  "deepseek/deepseek-chat",
35091
35091
  "deepseek/deepseek-reasoner",
35092
35092
  "free/deepseek-v4-flash",
35093
- "free/seed-oss-36b",
35094
35093
  "free/nemotron-3-nano-omni-30b-a3b-reasoning",
35095
35094
  "free/mistral-nemotron",
35096
35095
  "free/step-3.7-flash",
@@ -35400,9 +35399,10 @@ var init_models = __esm({
35400
35399
  "free/deepseek-v4-pro": "free/deepseek-v4-flash",
35401
35400
  "nvidia/deepseek-v4-flash": "free/deepseek-v4-flash",
35402
35401
  "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": "free/nemotron-3-nano-omni-30b-a3b-reasoning",
35403
- // qwen3-coder-480b retired (NVIDIA EOL 2026-06-14) → redirects to seed-oss-36b
35404
- // server-side; keep the explicit id mappings for pinned callers (BlockRun still
35405
- // resolves them), but point the generic coding shorthands at the live successor.
35402
+ // qwen3-coder-480b retired (NVIDIA EOL 2026-06-14). Its server-side redirect used
35403
+ // to be seed-oss-36b, but that EOL'd too on 2026-08-03, so blockrun re-pointed it
35404
+ // at gpt-oss-120b. Keep the explicit id mappings for pinned callers (BlockRun still
35405
+ // resolves them); the generic coding shorthands follow the gateway to gpt-oss-120b.
35406
35406
  "nvidia/qwen3-coder-480b": "free/qwen3-coder-480b",
35407
35407
  "qwen/qwen3-coder-480b-a35b-instruct": "free/qwen3-coder-480b",
35408
35408
  // glm-4.7 pin keeps the real id — blockrun redirects it server-side (→ gpt-oss-120b).
@@ -35448,13 +35448,14 @@ var init_models = __esm({
35448
35448
  "v4-flash": "free/deepseek-v4-flash",
35449
35449
  "mistral-free": "free/mistral-nemotron",
35450
35450
  // mistral-large-3-675b EOL'd 2026-07-28 → last live free Mistral (blockrun demoted its /free-mistral primary the same way)
35451
- "glm-free": "free/seed-oss-36b",
35452
- // qwen3-coder retired → live coder successor
35451
+ "glm-free": "free/gpt-oss-120b",
35452
+ // seed-oss-36b (the prior target) EOL'd 2026-08-03
35453
35453
  "llama-free": "free/gpt-oss-120b",
35454
35454
  // no live free Llama left (maverick died 2026-07)
35455
- "qwen-coder": "free/seed-oss-36b",
35456
- // qwen3-coder-480b retired → seed-oss-36b
35457
- "qwen-coder-free": "free/seed-oss-36b",
35455
+ // qwen3-coder-480b retired 2026-06-14 → seed-oss-36b, which then EOL'd 2026-08-03.
35456
+ // Follow the gateway's own retarget rather than chaining to a second dead model.
35457
+ "qwen-coder": "free/gpt-oss-120b",
35458
+ "qwen-coder-free": "free/gpt-oss-120b",
35458
35459
  "qwen-thinking": "free/gpt-oss-120b",
35459
35460
  // qwen3-next died 2026-07-17; no live free Qwen left
35460
35461
  "qwen3-next": "free/qwen3-next-80b-a3b-instruct",
@@ -35463,9 +35464,11 @@ var init_models = __esm({
35463
35464
  "mistral-small": "free/mistral-nemotron",
35464
35465
  // closest live Mistral-family free model
35465
35466
  // New live free models (2026-06-14 BlockRun free-tier refresh)
35467
+ // seed-oss pins name the model itself — kept routable, the gateway redirects them.
35466
35468
  "seed-oss": "free/seed-oss-36b",
35467
35469
  "seed-oss-36b": "free/seed-oss-36b",
35468
- "coder-free": "free/seed-oss-36b",
35470
+ "coder-free": "free/gpt-oss-120b",
35471
+ // generic "a free coder" → follows the gateway retarget
35469
35472
  "mistral-nemotron": "free/mistral-nemotron",
35470
35473
  "step-flash": "free/step-3.7-flash",
35471
35474
  "step-3.7-flash": "free/step-3.7-flash",
@@ -35486,8 +35489,9 @@ var init_models = __esm({
35486
35489
  "nemotron-super": "free/gpt-oss-120b",
35487
35490
  "nemotron-49b": "free/gpt-oss-120b",
35488
35491
  "nemotron-120b": "free/gpt-oss-120b",
35489
- devstral: "free/seed-oss-36b",
35490
- "devstral-2": "free/seed-oss-36b",
35492
+ devstral: "free/gpt-oss-120b",
35493
+ // seed-oss-36b EOL'd 2026-08-03; matches blockrun's own retarget
35494
+ "devstral-2": "free/gpt-oss-120b",
35491
35495
  maverick: "free/llama-4-maverick",
35492
35496
  // explicit-ish pin — gateway redirects
35493
35497
  free: "free/gpt-oss-120b",
@@ -36610,8 +36614,13 @@ var init_models = __esm({
36610
36614
  reasoning: true
36611
36615
  },
36612
36616
  {
36613
- // ByteDance Seed-OSS 36B: strong open-source coder, 131K context. Successor
36614
- // to the retired qwen3-coder-480b (BlockRun redirects qwen3-coder → seed-oss).
36617
+ // ByteDance Seed-OSS 36B: was the free coder, and the redirect target for the
36618
+ // retired qwen3-coder-480b. EOL'd 2026-08-03 — blockrun's probe got HTTP 410
36619
+ // Gone from NVIDIA on both passes and the prod health gate fired
36620
+ // [ALERT][free-model-dead] kind=gone the same day; upstream now hides it and
36621
+ // re-pointed its own dependents (qwen3-coder-480b, devstral-2) at gpt-oss-120b.
36622
+ // Entry kept so explicit pins stay routable; off the picker, the FREE_MODELS
36623
+ // cascade, and the router fallback chains.
36615
36624
  id: "free/seed-oss-36b",
36616
36625
  name: "[Free] Seed-OSS 36B",
36617
36626
  version: "oss-36b",
@@ -90216,8 +90225,6 @@ var init_proxy = __esm({
90216
90225
  "free/gpt-oss-20b",
90217
90226
  "free/deepseek-v4-flash",
90218
90227
  // 1M ctx; slow (~10 tok/s, 07-28 probe) but completes
90219
- "free/seed-oss-36b",
90220
- // live coder (successor to retired qwen3-coder-480b)
90221
90228
  "free/mistral-nemotron",
90222
90229
  // strong instruction following
90223
90230
  "free/step-3.7-flash",
package/dist/index.js CHANGED
@@ -93,7 +93,6 @@ var init_top_models = __esm({
93
93
  "deepseek/deepseek-chat",
94
94
  "deepseek/deepseek-reasoner",
95
95
  "free/deepseek-v4-flash",
96
- "free/seed-oss-36b",
97
96
  "free/nemotron-3-nano-omni-30b-a3b-reasoning",
98
97
  "free/mistral-nemotron",
99
98
  "free/step-3.7-flash",
@@ -403,9 +402,10 @@ var init_models = __esm({
403
402
  "free/deepseek-v4-pro": "free/deepseek-v4-flash",
404
403
  "nvidia/deepseek-v4-flash": "free/deepseek-v4-flash",
405
404
  "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning": "free/nemotron-3-nano-omni-30b-a3b-reasoning",
406
- // qwen3-coder-480b retired (NVIDIA EOL 2026-06-14) → redirects to seed-oss-36b
407
- // server-side; keep the explicit id mappings for pinned callers (BlockRun still
408
- // resolves them), but point the generic coding shorthands at the live successor.
405
+ // qwen3-coder-480b retired (NVIDIA EOL 2026-06-14). Its server-side redirect used
406
+ // to be seed-oss-36b, but that EOL'd too on 2026-08-03, so blockrun re-pointed it
407
+ // at gpt-oss-120b. Keep the explicit id mappings for pinned callers (BlockRun still
408
+ // resolves them); the generic coding shorthands follow the gateway to gpt-oss-120b.
409
409
  "nvidia/qwen3-coder-480b": "free/qwen3-coder-480b",
410
410
  "qwen/qwen3-coder-480b-a35b-instruct": "free/qwen3-coder-480b",
411
411
  // glm-4.7 pin keeps the real id — blockrun redirects it server-side (→ gpt-oss-120b).
@@ -451,13 +451,14 @@ var init_models = __esm({
451
451
  "v4-flash": "free/deepseek-v4-flash",
452
452
  "mistral-free": "free/mistral-nemotron",
453
453
  // mistral-large-3-675b EOL'd 2026-07-28 → last live free Mistral (blockrun demoted its /free-mistral primary the same way)
454
- "glm-free": "free/seed-oss-36b",
455
- // qwen3-coder retired → live coder successor
454
+ "glm-free": "free/gpt-oss-120b",
455
+ // seed-oss-36b (the prior target) EOL'd 2026-08-03
456
456
  "llama-free": "free/gpt-oss-120b",
457
457
  // no live free Llama left (maverick died 2026-07)
458
- "qwen-coder": "free/seed-oss-36b",
459
- // qwen3-coder-480b retired → seed-oss-36b
460
- "qwen-coder-free": "free/seed-oss-36b",
458
+ // qwen3-coder-480b retired 2026-06-14 → seed-oss-36b, which then EOL'd 2026-08-03.
459
+ // Follow the gateway's own retarget rather than chaining to a second dead model.
460
+ "qwen-coder": "free/gpt-oss-120b",
461
+ "qwen-coder-free": "free/gpt-oss-120b",
461
462
  "qwen-thinking": "free/gpt-oss-120b",
462
463
  // qwen3-next died 2026-07-17; no live free Qwen left
463
464
  "qwen3-next": "free/qwen3-next-80b-a3b-instruct",
@@ -466,9 +467,11 @@ var init_models = __esm({
466
467
  "mistral-small": "free/mistral-nemotron",
467
468
  // closest live Mistral-family free model
468
469
  // New live free models (2026-06-14 BlockRun free-tier refresh)
470
+ // seed-oss pins name the model itself — kept routable, the gateway redirects them.
469
471
  "seed-oss": "free/seed-oss-36b",
470
472
  "seed-oss-36b": "free/seed-oss-36b",
471
- "coder-free": "free/seed-oss-36b",
473
+ "coder-free": "free/gpt-oss-120b",
474
+ // generic "a free coder" → follows the gateway retarget
472
475
  "mistral-nemotron": "free/mistral-nemotron",
473
476
  "step-flash": "free/step-3.7-flash",
474
477
  "step-3.7-flash": "free/step-3.7-flash",
@@ -489,8 +492,9 @@ var init_models = __esm({
489
492
  "nemotron-super": "free/gpt-oss-120b",
490
493
  "nemotron-49b": "free/gpt-oss-120b",
491
494
  "nemotron-120b": "free/gpt-oss-120b",
492
- devstral: "free/seed-oss-36b",
493
- "devstral-2": "free/seed-oss-36b",
495
+ devstral: "free/gpt-oss-120b",
496
+ // seed-oss-36b EOL'd 2026-08-03; matches blockrun's own retarget
497
+ "devstral-2": "free/gpt-oss-120b",
494
498
  maverick: "free/llama-4-maverick",
495
499
  // explicit-ish pin — gateway redirects
496
500
  free: "free/gpt-oss-120b",
@@ -1613,8 +1617,13 @@ var init_models = __esm({
1613
1617
  reasoning: true
1614
1618
  },
1615
1619
  {
1616
- // ByteDance Seed-OSS 36B: strong open-source coder, 131K context. Successor
1617
- // to the retired qwen3-coder-480b (BlockRun redirects qwen3-coder → seed-oss).
1620
+ // ByteDance Seed-OSS 36B: was the free coder, and the redirect target for the
1621
+ // retired qwen3-coder-480b. EOL'd 2026-08-03 — blockrun's probe got HTTP 410
1622
+ // Gone from NVIDIA on both passes and the prod health gate fired
1623
+ // [ALERT][free-model-dead] kind=gone the same day; upstream now hides it and
1624
+ // re-pointed its own dependents (qwen3-coder-480b, devstral-2) at gpt-oss-120b.
1625
+ // Entry kept so explicit pins stay routable; off the picker, the FREE_MODELS
1626
+ // cascade, and the router fallback chains.
1618
1627
  id: "free/seed-oss-36b",
1619
1628
  name: "[Free] Seed-OSS 36B",
1620
1629
  version: "oss-36b",
@@ -36378,8 +36387,8 @@ var init_config = __esm({
36378
36387
  // FREE — smaller, faster
36379
36388
  "free/deepseek-v4-flash",
36380
36389
  // FREE — 1M ctx; slow (~10 tok/s, 07-28 probe) but completes
36381
- "free/seed-oss-36b",
36382
- // FREE — live coder (qwen3.5-122b + qwen3-next died 2026-07; mistral-large-3-675b EOL'd 07-28)
36390
+ // seed-oss-36b sat here as the free coder until it EOL'd 2026-08-03 (HTTP 410).
36391
+ // gpt-oss-120b/20b already head this chain, so the rung is dropped, not retargeted.
36383
36392
  "google/gemini-3.1-flash-lite",
36384
36393
  // $0.25/$1.50 — newest flash-lite
36385
36394
  "openai/gpt-5.4-nano",
@@ -36496,8 +36505,8 @@ var init_config = __esm({
36496
36505
  "openai/gpt-5.3-codex",
36497
36506
  "deepseek/deepseek-chat",
36498
36507
  // Cheap, reliable
36499
- "free/seed-oss-36b"
36500
- // NVIDIA free ultimate backstop (qwen3-coder-480b retired)
36508
+ "free/gpt-oss-120b"
36509
+ // NVIDIA free ultimate backstop (was seed-oss-36b; EOL'd 2026-08-03)
36501
36510
  ]
36502
36511
  },
36503
36512
  REASONING: {
@@ -36586,8 +36595,8 @@ var init_config = __esm({
36586
36595
  // Previous flagship — 6,213ms, reliable
36587
36596
  "deepseek/deepseek-chat",
36588
36597
  // 1,431ms — cheap, reliable
36589
- "free/seed-oss-36b"
36590
- // NVIDIA free ultimate backstop (qwen3-coder-480b retired)
36598
+ "free/gpt-oss-120b"
36599
+ // NVIDIA free ultimate backstop (was seed-oss-36b; EOL'd 2026-08-03)
36591
36600
  ]
36592
36601
  },
36593
36602
  REASONING: {
@@ -90036,8 +90045,6 @@ var init_proxy = __esm({
90036
90045
  "free/gpt-oss-20b",
90037
90046
  "free/deepseek-v4-flash",
90038
90047
  // 1M ctx; slow (~10 tok/s, 07-28 probe) but completes
90039
- "free/seed-oss-36b",
90040
- // live coder (successor to retired qwen3-coder-480b)
90041
90048
  "free/mistral-nemotron",
90042
90049
  // strong instruction following
90043
90050
  "free/step-3.7-flash",
package/docs/anthropic-cost-savings.md CHANGED
@@ -50,7 +50,7 @@ This is where you're paying for real value:
50
50
 
51
51
  ## The Solution: ClawRouter
52
52
 
53
- [ClawRouter](https://github.com/BlockRunAI/ClawRouter) is an open-source local proxy that sits between your app and <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> AI models. It saves you money in three ways: **smart routing**, **token optimization**, and **response caching**.
53
+ [ClawRouter](https://github.com/BlockRunAI/ClawRouter) is an open-source local proxy that sits between your app and <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> AI models. It saves you money in three ways: **smart routing**, **token optimization**, and **response caching**.
54
54
 
55
55
  ```
56
56
  ┌─────────────┐ ┌──────────────────────────────┐ ┌──────────────────┐
package/docs/anthropic-third-party-harness-changes.md CHANGED
@@ -31,7 +31,7 @@ This is exactly the problem [ClawRouter](https://github.com/BlockRunAI/ClawRoute
31
31
 
32
32
  ## ClawRouter: Smart Routing for Agents
33
33
 
34
- ClawRouter is an open-source local proxy that sits between your agent and <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> LLM models across 9 providers. It analyzes every request across 15 dimensions and routes it to the cheapest model that can handle it — in under 1ms, entirely locally.
34
+ ClawRouter is an open-source local proxy that sits between your agent and <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> LLM models across 9 providers. It analyzes every request across 15 dimensions and routes it to the cheapest model that can handle it — in under 1ms, entirely locally.
35
35
 
36
36
  ```
37
37
  Your Agent → ClawRouter (localhost:8402) → Best model for the job
@@ -74,7 +74,7 @@ Then point your agent at `http://localhost:8402/v1/` with any OpenAI-compatible
74
74
 
75
75
  - **No API keys to manage** — wallet-based cryptographic auth
76
76
  - **No subscriptions** — pay per request in USDC (Base or Solana)
77
- - **No vendor lock-in** — <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, switch anytime
77
+ - **No vendor lock-in** — <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, switch anytime
78
78
  - **You control your wallet** — non-custodial, funds never held by a third party
79
79
  - **Budget caps** — set a max spend per session, ClawRouter gracefully downgrades when budget runs low
80
80
  - **Token compression** — 7-layer pipeline reduces token costs by 7–40% before they hit any provider
package/docs/clawrouter-cuts-llm-api-costs-500x.md CHANGED
@@ -135,7 +135,7 @@ Less context repeated = fewer tokens = lower cost. Agents that need to recall ea
135
135
 
136
136
  ### 7. x402 Micropayments — Wallet-Based Budget Control
137
137
 
138
- <p align="center"><img src="assets/blockrun-clawrouter-x402-usdc-micropayment-wallet-budget-control.png" alt="Budget limits enforced by physical construction — wallet loaded via Base/Solana, pay per call across <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, balance hits zero and the valve shuts cleanly" width="720"></p>
138
+ <p align="center"><img src="assets/blockrun-clawrouter-x402-usdc-micropayment-wallet-budget-control.png" alt="Budget limits enforced by physical construction — wallet loaded via Base/Solana, pay per call across <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, balance hits zero and the valve shuts cleanly" width="720"></p>
139
139
 
140
140
  ClawRouter pays for inference via [x402](https://x402.org/) USDC micropayments (Base or Solana). You load a wallet. Each inference call costs exactly what it costs. When the wallet runs low, requests stop cleanly.
141
141
 
package/docs/clawrouter-vs-openrouter-llm-routing-comparison.md CHANGED
@@ -222,7 +222,7 @@ OpenRouter doesn't always pass through provider-specific features correctly. Ima
222
222
 
223
223
  **Direct provider routing.** ClawRouter routes through BlockRun's API directly to providers — not through a second aggregator. One hop, not two. Provider-specific features work because there's no middleman translating them.
224
224
 
225
- ![Guaranteed Feature Parity & Direct Connectivity — Three-panel diagram: Vision (image_url auto-detected → vision-capable models only), Tool Calling (toolCalling flag → agentic models only), Catalog (curated <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models with automatic legacy-to-modern redirects). Direct provider routing means no dropped payloads.](./assets/clawrouter-feature-parity-direct-connectivity.png)
225
+ ![Guaranteed Feature Parity & Direct Connectivity — Three-panel diagram: Vision (image_url auto-detected → vision-capable models only), Tool Calling (toolCalling flag → agentic models only), Catalog (curated <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models with automatic legacy-to-modern redirects). Direct provider routing means no dropped payloads.](./assets/clawrouter-feature-parity-direct-connectivity.png)
226
226
 
227
227
  ---
228
228
 
@@ -240,7 +240,7 @@ When new models launch, OpenRouter's catalog lags. Users configure a model that
240
240
 
241
241
  ### How ClawRouter Solves This
242
242
 
243
- ClawRouter maintains a curated catalog of <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models across 9 providers (including 9 free models), updated with each release. Delisted models have automatic redirect aliases:
243
+ ClawRouter maintains a curated catalog of <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models across 9 providers (including 9 free models), updated with each release. Delisted models have automatic redirect aliases:
244
244
 
245
245
  ```typescript
246
246
  // Delisted models redirect automatically
@@ -268,7 +268,7 @@ No silent drops. No stale catalog. Models are benchmarked for speed, quality, an
268
268
  | **Rate limits** | Per-key, shared | Per-wallet, independent |
269
269
  | **Vision support** | Images sometimes dropped | Auto-detected, vision-only fallback |
270
270
  | **Tool calling** | Silent failures with some models | Flag-based filtering, guaranteed support |
271
- | **Model catalog** | Laggy, silent drops | Curated <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models, redirect aliases |
271
+ | **Model catalog** | Laggy, silent drops | Curated <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models, redirect aliases |
272
272
  | **Budget control** | Monthly invoice | Per-session cap (`maxCostPerRun`) |
273
273
  | **Setup** | Create account, paste key | Agent generates wallet, auto-configured |
274
274
  | **Average cost** | $25/M tokens (Opus direct) | auto-routed = **<!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->% savings** |
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@blockrun/clawrouter",
3
- "version": "0.12.240",
4
- "description": "Smart LLM router — save 87% on inference costs. 65 models (7 free), one wallet, x402 micropayments.",
3
+ "version": "0.12.241",
4
+ "description": "Smart LLM router — save 87% on inference costs. 71 models (6 free), one wallet, x402 micropayments.",
5
5
  "type": "module",
6
6
  "main": "dist/index.js",
7
7
  "types": "dist/index.d.ts",
package/skills/clawrouter/SKILL.md CHANGED
@@ -1,6 +1,6 @@
1
1
  ---
2
2
  name: clawrouter
3
- description: Hosted-gateway LLM router — save 87% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 65 models from OpenAI, Anthropic, Google, DeepSeek, xAI, NVIDIA, and more. 7 free NVIDIA models included. Also exposes realtime market data (global stocks, crypto, FX, commodities), Twitter/X intelligence, prediction-market data across Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow + UMA oracle resolution + wallet identity & clustering, phone-number intelligence (carrier + SIM-swap fraud detection) plus AI-powered outbound voice calls (Twilio + Bland.ai), AND the Surf unified crypto data API (84 endpoints — CEX/DEX, on-chain SQL over 80+ ClickHouse tables, 100M+ labeled wallets, prediction markets, social/CT mindshare, news, VC fund intel) as built-in agent tools. Not a local-inference tool — prompts are sent to the blockrun.ai gateway.
3
+ description: Hosted-gateway LLM router — save 87% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 71 models from OpenAI, Anthropic, Google, DeepSeek, xAI, NVIDIA, and more. 6 free NVIDIA models included. Also exposes realtime market data (global stocks, crypto, FX, commodities), Twitter/X intelligence, prediction-market data across Polymarket, Kalshi, Limitless, Opinion, Predict.Fun, dFlow + UMA oracle resolution + wallet identity & clustering, phone-number intelligence (carrier + SIM-swap fraud detection) plus AI-powered outbound voice calls (Twilio + Bland.ai), AND the Surf unified crypto data API (84 endpoints — CEX/DEX, on-chain SQL over 80+ ClickHouse tables, 100M+ labeled wallets, prediction markets, social/CT mindshare, news, VC fund intel) as built-in agent tools. Not a local-inference tool — prompts are sent to the blockrun.ai gateway.
4
4
  triggers:
5
5
  - "clawrouter"
6
6
  - "claw router"
@@ -53,7 +53,7 @@ metadata:
53
53
 
54
54
  # ClawRouter
55
55
 
56
- Hosted-gateway LLM router that saves <!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->% on inference costs by forwarding each request to the blockrun.ai gateway, which picks the cheapest model capable of handling it across <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models from 9 providers (<!-- br:models.free -->7<!-- /br:models.free --> free NVIDIA models). All billing flows through one USDC wallet; you do not hold provider API keys.
56
+ Hosted-gateway LLM router that saves <!-- br:savings.autoVsBaselinePct -->87<!-- /br:savings.autoVsBaselinePct -->% on inference costs by forwarding each request to the blockrun.ai gateway, which picks the cheapest model capable of handling it across <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models from 9 providers (<!-- br:models.free -->6<!-- /br:models.free --> free NVIDIA models). All billing flows through one USDC wallet; you do not hold provider API keys.
57
57
 
58
58
  **This is not a local-inference tool.** ClawRouter is a thin local proxy. Your prompts are sent over HTTPS to the blockrun.ai gateway for model execution. If your workload requires inference that never leaves your machine, use a local runtime like Ollama — ClawRouter is not the right tool for that use case.
59
59
 
@@ -141,7 +141,7 @@ Rules handle ~~80% of requests in <1ms. Only ambiguous queries hit the LLM class
141
141
 
142
142
  ## Available Models
143
143
 
144
- <!-- br:models.chatVisible -->65<!-- /br:models.chatVisible --> models including: gpt-5.6-terra [balanced, stable default], gpt-5.6-sol [flagship], gpt-5.6-luna [cost-efficient], gpt-5.5, gpt-5.5-pro [max compute], chat-latest [ChatGPT Instant], gpt-5.4, gpt-4o, o3, claude-fable-5, claude-opus-5 [Anthropic flagship, 1M ctx], claude-opus-4.8, claude-opus-4.7, claude-opus-4.5, claude-sonnet-5, claude-sonnet-4.6, gemini-3.1-pro, gemini-2.5-flash, deepseek-v4-pro, deepseek-chat, grok-4.5, grok-4.3, grok-build-0.1, kimi-k3 [1M ctx flagship], kimi-k2.7, qwen3.7-max [Qwen flagship, 1M ctx], and 7 free NVIDIA models (deepseek-v4-flash [1M context], seed-oss-36b [coding], nemotron-3-nano-omni-30b-a3b-reasoning [vision], mistral-nemotron, step-3.7-flash, nemotron-nano-9b-v2, nemotron-nano-12b-v2-vl [vision]).
144
+ <!-- br:models.chatVisible -->71<!-- /br:models.chatVisible --> models including: gpt-5.6-terra [balanced, stable default], gpt-5.6-sol [flagship], gpt-5.6-luna [cost-efficient], gpt-5.5, gpt-5.5-pro [max compute], chat-latest [ChatGPT Instant], gpt-5.4, gpt-4o, o3, claude-fable-5, claude-opus-5 [Anthropic flagship, 1M ctx], claude-opus-4.8, claude-opus-4.7, claude-opus-4.5, claude-sonnet-5, claude-sonnet-4.6, gemini-3.1-pro, gemini-2.5-flash, deepseek-v4-pro, deepseek-chat, grok-4.5, grok-4.3, grok-build-0.1, kimi-k3 [1M ctx flagship], kimi-k2.7, qwen3.7-max [Qwen flagship, 1M ctx], and 6 free NVIDIA models (deepseek-v4-flash [1M context], nemotron-3-nano-omni-30b-a3b-reasoning [vision], mistral-nemotron, step-3.7-flash, nemotron-nano-9b-v2, nemotron-nano-12b-v2-vl [vision]).
145
145
 
146
146
  ## Built-in Agent Tools
147
147
 
package/src/top-models.json CHANGED
@@ -39,7 +39,6 @@
39
39
  "deepseek/deepseek-chat",
40
40
  "deepseek/deepseek-reasoner",
41
41
  "free/deepseek-v4-flash",
42
- "free/seed-oss-36b",
43
42
  "free/nemotron-3-nano-omni-30b-a3b-reasoning",
44
43
  "free/mistral-nemotron",
45
44
  "free/step-3.7-flash",