Model Scan Report — 2026-03-14

Generated 2026-03-14 03:30 EDT

Master Specs Database

Loading OpenClaw pi-ai catalog... ✓ 790 models from pi-ai catalog (cached to piai-catalog.json) Fetching LiteLLM specs... ✓ 2582 entries cached to litellm-specs.json Fetching OpenRouter catalog... ✓ 344 models cached to openrouter-specs.json

✅ Master specs: 2709 models (+11 vs last run) Layer 1 — pi-ai catalog: 790 models (OpenClaw built-ins) Layer 2 — LiteLLM: 2582 entries Layer 2 — OpenRouter: 344 models With cost data: 1946 models

anthropic

⚠️ Failed to fetch (network error, auth issue, or balance depleted)

openai

No changes. (125 models)

google

No changes. (45 models)

groq

No changes. (18 models)

mistral

🆕 New (4)

mistral-squarepoint-2602
voxtral-mini-realtime-2602
voxtral-mini-realtime-latest
voxtral-mini-transcribe-realtime-2602

cerebras

🆕 New (1)

qwen-3-235b-a22b-instruct-2507

❌ Removed (1)

gpt-oss-120b

xai

⚠️ Failed to fetch (network error, auth issue, or balance depleted)

deepseek

No changes. (2 models)

dashscope

No changes. (168 models)

nvidia

No changes. (188 models)

together

⚠️ Could not parse response.

sambanova

No changes. (16 models)

codestral

⚠️ Failed to fetch (network error, auth issue, or balance depleted)

zai

No changes. (5 models)

Catalog Sync

📋 Catalog: 990 models across 17 providers

• cerebras: +1 (qwen-3-235b-a22b-instruct-2507)
• cerebras: -1 removed
• dashscope: +134 (MiniMax-M2.5, deepseek-r1-distill-llama-70b, deepseek-r1-distill-llama-8b…)
• huggingface: +6 (CohereLabs/command-a-reasoning-08-2025, Qwen/Qwen2.5-VL-7B-Instruct, Sao10K/L3-70B-Euryale-v2.1…)
• huggingface: -7 removed
• mistral: +1 (mistral-squarepoint-2602)
• nvidia: +160 (01-ai/yi-large, abacusai/dracarys-llama-3.1-70b-instruct, adept/fuyu-8b…)
• openrouter: +334 (ai21/jamba-large-1.7, aion-labs/aion-1.0, aion-labs/aion-1.0-mini…)
• openrouter: -3 removed
• sambanova: +14 (DeepSeek-R1-0528, DeepSeek-R1-Distill-Llama-70B, DeepSeek-V3.1…)
• zai: +4 (glm-4.5, glm-4.5-air, glm-4.6…)
• zai: -1 removed

Net: +654 added, -12 removed

Config Audit

✅ Auto-Replaced (3)

• cerebras/gpt-oss-120b → cerebras/qwen-3-235b-a22b-instruct-2507 (strategy: same-provider) @ config.agents.defaults.model.fallbacks[0]
• cerebras/gpt-oss-120b → cerebras/qwen-3-235b-a22b-instruct-2507 (strategy: same-provider) @ config.agents.list[1].model.fallbacks[1]
• cerebras/gpt-oss-120b → cerebras/qwen-3-235b-a22b-instruct-2507 (strategy: same-provider) @ config.agents.list[6].model.fallbacks[1]

Config Validation

🔍 Incomplete Model Entries (3)

• together/Qwen/Qwen2.5-Coder-32B-Instruct missing: input, contextWindow, maxTokens
• together/deepseek-ai/DeepSeek-R1-Distill-Llama-70B missing: input, contextWindow, maxTokens
• codestral/codestral-latest missing: input, contextWindow, maxTokens

Final state: 990 models | 17 providers in catalog | 7 providers configured in openclaw.json

---

Summary

• Providers scanned: 14
• 9 change(s)/issue(s) — review recommended