Issue Station
Sign In
Tools hereToolbox

hermes - 💡(How to fix) Fix feat: MoA customizável com multi-key NVIDIA NIM (interesse em PR?) [1 participants]

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…
GitHub stats
NousResearch/hermes-agent#15101•Fetched 2026-04-25 06:24:34
View on GitHub
Comments
0
Participants
1
Timeline
4
Reactions
0
Author
MestreY0d4-UninterMestreY0d4-Uninter
Participants
MestreY0d4-UninterMestreY0d4-Uninter
Timeline (top)
labeled ×4
RAW_BUFFERClick to expand / collapse

MoA com NVIDIA NIM gratuito — branch disponível

Implementamos uma extensão do Mixture of Agents que roteia as chamadas de referência e agregação pelo NVIDIA NIM (tier gratuito), com suporte a múltiplas API keys em pool rotativo.

O que foi feito

FeatureDescrição
Pool multi-key NVIDIANVIDIA_API_KEY até NVIDIA_API_KEY_4 com round-robin automático
TTL floor per-providerFloor de 180s para NVIDIA após 429, evita bounce em key ainda esgotada
fallback_providerSe o provider primário falhar, tenta um secundário configurável
Defaults otimizadosLineup kimi-k1.5 + nemotron-70b + maverick → qwen480b (resultado de ablação)
Setup/status provider-awarehermes tools não mente mais sobre qual provider o MoA usa
Docs em AGENTS.mdSeção completa sobre escolha de provider, multi-key e fallback

Motivação

O tier gratuito do NVIDIA NIM tem rate-limit agressivo por key. Com uma só key o MoA fica ocioso na maior parte do tempo. Com pool de 4 keys o throughput real sobe ~4x sem custo.

Branch

O código está em um repositório privado por enquanto. Se houver interesse em revisar e incorporar, abro PR — só queria confirmar antes se faz sentido para a direção do projeto.

13 commits, 64 testes cobrindo os novos caminhos, zero regressões nos testes do repositório.

extent analysis

TL;DR

Review and test the new branch with NVIDIA NIM integration to ensure its stability and performance before merging it into the main project.

Guidance

  • Verify the functionality of the round-robin API key rotation and the TTL floor implementation to ensure they are working as expected.
  • Test the fallback_provider feature to confirm it correctly switches to a secondary provider when the primary one fails.
  • Review the updated documentation in AGENTS.md to ensure it accurately reflects the new features and configuration options.
  • Evaluate the performance benefits of using a pool of 4 API keys and the optimized defaults.

Example

No code snippet is provided as the issue does not contain specific code examples.

Notes

The issue lacks information about potential conflicts or compatibility issues with the existing codebase, so thorough testing is recommended.

Recommendation

Apply workaround: Review and test the new branch before merging to ensure its stability and performance, as the issue seems to be a feature implementation rather than a bug fix.

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…
#api#logging issue#authentication issue#prompt issue#agent setup

Still need to ship something?

×6

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

Back to top recommendations

TRENDING