Agents Directory
SkillsRankingsAgents
CategoriesModelsBenchmarksCompareAgent LeaderboardSkillsRankingsAgentsAbout
/Rankings
/Best cheap models for Hermes
Hermes logo

Best cheap models for Hermes

Cheap

Hermes runs all day, so the model price is your real subscription fee. These are the best cheap models for Hermes right now: everything costs at most about $0.75 per million input tokens, ranked by how much agent capability survives the price cut.

Which model should you run?

As of June 2026, the cheap-model question for Hermes logoHermes splits on how you pay:

  • Have a ChatGPT Codex logoCodex subscription? Run OpenAIGPT-5.4 Mini on your plan and pay nothing extra per token
  • Paying per token? Route DeepSeekDeepSeek V4 or GeminiGemini 3 Flash through OpenRouter with one key
  • Running Hermes logoHermes always-on or with sub-agents? DeepSeekDeepSeek V4 Flash or GLM 4.7 Flash logoGLM 4.7 Flash keep a full day of runs at cents

Whatever you pick, keep one flagship configured for the hard tasks. The cheap picks handle the everyday 90% and slip on long multi-step planning.

The best cheap model for Hermes logoHermes is DeepSeekDeepSeek V4: near-frontier quality, a 1M token context window, and open weights at $0.435 per million input tokens, all through one OpenRouter key.

GeminiGemini 3 Flash at $0.50 per million is the fastest cheap pick for high-volume runs, and OpenAIGPT-5.4 Mini at $0.75 makes sense if you already have a ChatGPT Codex logoCodex subscription, since it runs on your plan instead of per token.

For always-on background work and sub-agents, drop to DeepSeekDeepSeek V4 Flash ($0.098) or GLM 4.7 Flash logoGLM 4.7 Flash ($0.06). At those prices a full day of routine Hermes logoHermes runs costs cents, not dollars.

We also have more in-depth Hermes rankings:Best free models for HermesBest models for HermesBest open-source models for Hermes
The ranking
Updated June 2026
Best cheap daily drivers

Strong enough to be your main Hermes model, at a fraction of flagship prices.

#ModelContextInput
  • 1
    1DeepSeek
    DeepSeek V4DeepSeekThe best cheap default: near-frontier quality, 1M context, $0.435 per million.
    Context1.049M
    Input$0.435/M
  • 2
    2Gemini
    Gemini 3 FlashGoogle DeepMindThe fastest cheap pick for high-volume runs, $0.50 per million.
    Context1.049M
    Input$0.5/M
  • 3
    3OpenAI
    GPT-5.4 MiniOpenAIThe cheapest GPT that still handles real agent work. Codex subscription or API.
    Context400K
    Input$0.75/M
  • 4
    4MoonshotAI
    Kimi K2.5Moonshot AIOpen-weight agentic coder at $0.40 per million input tokens.
    Context262K
    Input$0.375/M
  • 5
    5Minimax
    MiniMax-M3MiniMaxNear-frontier intelligence at $0.30 per million input tokens.
    Context205K
    Input$0.3/M
Best for always-on and sub-agents

When Hermes runs continuously or fans out to sub-agents, these prices make it painless.

#ModelContextInput
  • 1
    1DeepSeek
    DeepSeek V4 FlashDeepSeekA 1M context window at $0.098 per million, for always-on background work.
    Context1.049M
    Input$0.09/M
  • 2
    2GLM 4.7 Flash logo
    GLM 4.7 FlashZ.AIThe floor: $0.06 per million for sub-agents and routine steps.
    Context203K
    Input$0.06/M
Intelligence vs. price

Each model's Artificial Analysis Intelligence Index logoArtificial Analysis Intelligence Index score against its blended price per 1M tokens. Toward the top right is more intelligence per dollar.

Full interactive leaderboard on our Intelligence Index page.

Frequently asked questions
What is the cheapest good model for Hermes?

DeepSeekDeepSeek V4 is the best cheap model for Hermes logoHermes: near-frontier quality with a 1M token context window at $0.435 per million input tokens via OpenRouter. If you want the absolute floor that still works, GLM 4.7 Flash logoGLM 4.7 Flash at $0.06 per million handles routine steps and sub-agents for cents per day.

How much does it cost to run Hermes on a cheap model?

A typical active day of Hermes logoHermes use runs tens of millions of tokens. On DeepSeekDeepSeek V4 at $0.435 per million input tokens, that lands around a dollar or two per day; on DeepSeekDeepSeek V4 Flash or GLM 4.7 Flash logoGLM 4.7 Flash it drops to cents. The same day on a flagship like ClaudeClaude Opus 4.8 or OpenAIGPT-5.5 can cost ten times more, which is why most Hermes logoHermes users run a cheap default and escalate selectively.

Should I use a cheap model or a free model for Hermes?

Free models on OpenRouter work for trying Hermes logoHermes out, but rate limits bite quickly on an agent that runs all day. The cheap picks here remove that ceiling for a few dollars a month of typical use. A common setup: a cheap model as your default, free models for experiments, and one flagship for hard tasks. See our best free models for Hermes logoHermes ranking for the $0 options.

Do cheap models work with Hermes skills and tools?

Yes. DeepSeekDeepSeek V4, GeminiGemini 3 Flash, OpenAIGPT-5.4 Mini, MoonshotAIKimi K2.5, and MinimaxMiniMax-M3 all handle tool calling and skill use reliably for everyday tasks. The gap versus flagships shows up in long multi-step plans, so route those to a stronger model rather than abandoning the cheap default.

More rankings for Hermes
Hermes logo

Best free models for Hermes

Want to run Hermes without paying per token? OpenRouter serves plenty of capable models at $0 (rate limits apply). These are the best free models for Hermes, grouped by what you run it for: general agent work, coding, and fast high-volume steps.
Hermes logo

Best models for Hermes

Hermes runs your skills locally and leans on the model for planning and skill use. These are the models that pair best with it right now, grouped by what you actually want to spend and ranked on real agentic-coding performance.
Hermes logo

Best open-source models for Hermes

Open-weight models you can inspect, fine-tune, and self-host. Ideal for privacy-sensitive or air-gapped Hermes setups. We rank a deep bench here because Hermes routes more than 380 different models in the wild, and most of its real volume goes to open weights.
Share:
Details:

  • Agent

    Hermes logoHermes

  • Models

    7

  • Filter

    Cheap

  • Updated

    June 2026
Browse:SkillsRankingsModelsBenchmarksProvidersAgentsAgent LeaderboardCompareCategories
Quick Links:AboutBlog

© 2026 Agents Directory