Issue Station
Sign In
Tools hereToolbox

openclaw - 💡(How to fix) Fix [Bug]: Failover selects unconfigured model MiniMax-M2.7-highspeed causing complete session failure

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

When minimax-portal/MiniMax-M2.7 receives a rate_limit_error, OpenClaw's failover selects MiniMax-M2.7-highspeed as fallback — a model that is neither declared in openclaw.json nor available in the Token Plan Starter subscription. It immediately hits the same rate limit, causing complete session failure with FailoverError.

Error Message

01:59:02 [agent/embedded] embedded run agent end: runId=commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee isError=true model=MiniMax-M2.7-highspeed provider=minimax-portal error=⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056) rawError={"type":"error","error":{"type":"rate_limit_error","message":"usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"}}
01:59:02 [agent/embedded] embedded run failover decision: runId=commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee stage=assistant decision=fallback_model reason=rate_limit from=minimax-portal/MiniMax-M2.7-highspeed profile=sha256:9e08bd6be9c1
01:59:02 [diagnostic] lane task error: lane=main durationMs=8170 error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [diagnostic] lane task error: lane=session:agent:dc-channel--1476858065914695741:commitments:commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee durationMs=8176 error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [commitments] commitment extraction failed

Root Cause

When minimax-portal/MiniMax-M2.7 receives a rate_limit_error, OpenClaw's failover selects MiniMax-M2.7-highspeed as fallback — a model that is neither declared in openclaw.json nor available in the Token Plan Starter subscription. It immediately hits the same rate limit, causing complete session failure with FailoverError.

Code Example

1. OpenClaw 2026.5.26 running on Windows_NT with minimax-portal/MiniMax-M2.7 as primary model.
2. Primary model hits Token Plan Starter hourly rate limit (resets at 2026-05-28T20:00:00Z).
3. Failover logic selects MiniMax-M2.7-highspeed during `decision=fallback_model reason=rate_limit`.
4. MiniMax-M2.7-highspeed is not in openclaw.json config — it appears to be auto-generated by OpenClaw internally.
5. Rate limit is hit on this unconfigured model, exhausting all fallback attempts.
6. Session fails with: FailoverError — "usage limit exceeded, 5-hour usage limit reached".

---

01:59:02 [agent/embedded] embedded run agent end: runId=... isError=true model=MiniMax-M2.7-highspeed provider=minimax-portal error=⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used)
01:59:02 [agent/embedded] embedded run failover decision: runId=... stage=assistant decision=fallback_model reason=rate_limit from=minimax-portal/MiniMax-M2.7-highspeed profile=sha256:9e08bd6be9c1
01:59:02 [diagnostic] lane task error: lane=main error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [commitments] commitment extraction failed

---

{
  "models.providers.minimax-portal": {
    "baseUrl": "https://api.minimax.io/anthropic",
    "api": "anthropic-messages",
    "models": [
      { "id": "MiniMax-M2.5" },
      { "id": "MiniMax-M2.5-Lightning" },
      { "id": "MiniMax-M2.7" },
      { "id": "MiniMax-M2.1" },
      { "id": "MiniMax-VL-01" }
    ]
  }
}

---

{
  "agents.defaults.model": {
    "primary": "minimax-portal/MiniMax-M2.7",
    "fallbacks": [
      "minimax-portal/MiniMax-M2.5",
      "minimax-portal/MiniMax-M2.1"
    ]
  }
}

---

01:59:02 [agent/embedded] embedded run agent end: runId=commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee isError=true model=MiniMax-M2.7-highspeed provider=minimax-portal error=⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056) rawError={"type":"error","error":{"type":"rate_limit_error","message":"usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"}}
01:59:02 [agent/embedded] embedded run failover decision: runId=commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee stage=assistant decision=fallback_model reason=rate_limit from=minimax-portal/MiniMax-M2.7-highspeed profile=sha256:9e08bd6be9c1
01:59:02 [diagnostic] lane task error: lane=main durationMs=8170 error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [diagnostic] lane task error: lane=session:agent:dc-channel--1476858065914695741:commitments:commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee durationMs=8176 error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [commitments] commitment extraction failed
RAW_BUFFERClick to expand / collapse

Bug type: Behavior bug (incorrect output/state without crash)

Beta release blocker: No

Summary

When minimax-portal/MiniMax-M2.7 receives a rate_limit_error, OpenClaw's failover selects MiniMax-M2.7-highspeed as fallback — a model that is neither declared in openclaw.json nor available in the Token Plan Starter subscription. It immediately hits the same rate limit, causing complete session failure with FailoverError.

Steps to reproduce

1. OpenClaw 2026.5.26 running on Windows_NT with minimax-portal/MiniMax-M2.7 as primary model.
2. Primary model hits Token Plan Starter hourly rate limit (resets at 2026-05-28T20:00:00Z).
3. Failover logic selects MiniMax-M2.7-highspeed during `decision=fallback_model reason=rate_limit`.
4. MiniMax-M2.7-highspeed is not in openclaw.json config — it appears to be auto-generated by OpenClaw internally.
5. Rate limit is hit on this unconfigured model, exhausting all fallback attempts.
6. Session fails with: FailoverError — "usage limit exceeded, 5-hour usage limit reached".

Expected behavior

  1. Failover should only select models present in openclaw.json config AND available in the subscription.
  2. If no valid fallback exists, fail with a clear "no valid fallback model" error message — not by attempting an undeclared model.
  3. The model name MiniMax-M2.7-highspeed should not be generated or called without explicit config.

Actual behavior

01:59:02 [agent/embedded] embedded run agent end: runId=... isError=true model=MiniMax-M2.7-highspeed provider=minimax-portal error=⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used)
01:59:02 [agent/embedded] embedded run failover decision: runId=... stage=assistant decision=fallback_model reason=rate_limit from=minimax-portal/MiniMax-M2.7-highspeed profile=sha256:9e08bd6be9c1
01:59:02 [diagnostic] lane task error: lane=main error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [commitments] commitment extraction failed

OpenClaw version

2026.5.26

Operating system

Windows_NT 10.0.26200 (x64)

Install method

npm global

Model

minimax-portal/MiniMax-M2.7 (primary) Configured fallbacks: minimax-portal/MiniMax-M2.5, minimax-portal/MiniMax-M2.1 No MiniMax-M2.7-highspeed exists in any config file.

Provider / routing chain

minimax-portal (direct, no intermediate gateway)

Additional provider/model setup details

Relevant models.providers.minimax-portal config declares only these models — MiniMax-M2.7-highspeed does not appear:

{
  "models.providers.minimax-portal": {
    "baseUrl": "https://api.minimax.io/anthropic",
    "api": "anthropic-messages",
    "models": [
      { "id": "MiniMax-M2.5" },
      { "id": "MiniMax-M2.5-Lightning" },
      { "id": "MiniMax-M2.7" },
      { "id": "MiniMax-M2.1" },
      { "id": "MiniMax-VL-01" }
    ]
  }
}
{
  "agents.defaults.model": {
    "primary": "minimax-portal/MiniMax-M2.7",
    "fallbacks": [
      "minimax-portal/MiniMax-M2.5",
      "minimax-portal/MiniMax-M2.1"
    ]
  }
}

Auth profile: minimax-portal:default mode=oauth.

Logs

01:59:02 [agent/embedded] embedded run agent end: runId=commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee isError=true model=MiniMax-M2.7-highspeed provider=minimax-portal error=⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056) rawError={"type":"error","error":{"type":"rate_limit_error","message":"usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"}}
01:59:02 [agent/embedded] embedded run failover decision: runId=commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee stage=assistant decision=fallback_model reason=rate_limit from=minimax-portal/MiniMax-M2.7-highspeed profile=sha256:9e08bd6be9c1
01:59:02 [diagnostic] lane task error: lane=main durationMs=8170 error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [diagnostic] lane task error: lane=session:agent:dc-channel--1476858065914695741:commitments:commitments-a6708113-1738-4a41-8e8e-aa48dafe14ee durationMs=8176 error="FailoverError: ⚠️ usage limit exceeded, 5-hour usage limit reached for Token Plan Starter (0/0 used), resets at 2026-05-28T20:00:00Z (2056)"
01:59:02 [commitments] commitment extraction failed

Impact and severity

  • Affected: Discord #ai-小助理 channel, any minimax-portal session during rate-limit window.
  • Severity: High (blocks workflow — session must fail entirely).
  • Frequency: Every time primary model hits rate limit while MiniMax-M2.7-highspeed is auto-generated as fallback.
  • Consequence: Failover attempts an unconfigured auto-generated model name, wasting API error budget and masking the real problem (no valid fallback available for the subscription tier).

Additional information

The model name MiniMax-M2.7-highspeed is not present in:

  • openclaw.json (grep -i highspeed returns no match)
  • MiniMax portal Token Plan Starter subscription
  • Any explicit profile or alias config

Its appearance in profile=sha256:9e08bd6be9c1 suggests OpenClaw's failover logic is internally deriving or constructing a model name from the provider catalog or a profile hash, without verifying that the model is explicitly configured or subscribed.

Suggested fix direction: failover logic should restrict fallback candidates to models explicitly declared in openclaw.json and confirmed available in the active subscription plan.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

  1. Failover should only select models present in openclaw.json config AND available in the subscription.
  2. If no valid fallback exists, fail with a clear "no valid fallback model" error message — not by attempting an undeclared model.
  3. The model name MiniMax-M2.7-highspeed should not be generated or called without explicit config.

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING