Standard Compute
← All models

Best Model for OpenClaw

Claude Sonnet 5.5 is the best primary model for OpenClaw: it scores 61.8% on Terminal-Bench 4.0, close to Claude Opus 5.5's 64.9%, at half the price ($5.33 per 100 coding-agent requests on our measured traffic). OpenClaw's docs advise setting the primary to the strongest latest-generation model you can use and avoiding older or weaker tiers for tool-enabled agents and untrusted inputs, which matters for an agent that reads your messages and the web. The fallback list then handles cost and outages: a model from a second provider, such as GPT-6.1 Sol, keeps the agent running when one provider fails.

Read the full OpenClaw review → · updated 2026-10-17

Sources: OpenClaw model docs · Terminal-Bench 4.0 leaderboard · Anthropic on Haiku 5.5 · DeepSeek V4.1 Flash release · Cost per request methodology

Run any of these models in your agent

Or let the router send each step to the cheapest model that can do it, on providers we vet for price, uptime and speed. Plans from $19 a month.

OpenClaw setup →See plans
#1Claude Sonnet 5.5Anthropic · $4.40/1M · 1M ctx

Best overall. The primary to start with: 61.8% on Terminal-Bench 4.0 at $2/$10 per million tokens, with a 1M-token context for long skill chains.

Overall 96/100 · Value 59Full breakdown →
#2Claude Opus 5.5Anthropic · $8.80/1M · 1M ctx

Best for long agent runs. #1 on Terminal-Bench 4.0 at 64.9%. Use it as the primary for long multi-step automations, or for agents that handle untrusted input.

Overall 98/100 · Value 51Full breakdown →
#3GPT-6.1 SolOpenAI · $4.40/1M · 1.1M ctx

Fallback from a second provider. 58.2% on Terminal-Bench 4.0 at the same $2/$10 list price as Sonnet 5.5, so a provider outage doesn't drop the agent to a much weaker model.

Overall 92/100 · Value 58Full breakdown →
#4Claude Haiku 5.5Anthropic · $0.22/1M · 1M ctx

Best value. Anthropic reports 39.2% on Terminal-Bench 4.0 at $0.10/$0.50 per million tokens. A sensible fallback for quick, low-stakes requests; prompts over 100K tokens cost five times as much.

Overall 82/100 · Value 97Full breakdown →
#5DeepSeek V4.1 FlashDeepSeek · $0.22/1M · 1.0M ctx

Budget pick. Open weights at $0.30/$1.20 per million tokens at peak hours and half off-peak. DeepSeek reports 42.3% on Terminal-Bench 4.0 for its October 17 update, up from 31.2%, so keep it away from tool-enabled agents that read untrusted content.

Overall 77/100 · Value 93Full breakdown →
#6GLM 5.3Z.AI · $1.50/1M · 1.0M ctx

Open-weight pick. The top open-weight model on Terminal-Bench 4.0 (41.8%), for local-first setups that self-host their primary model.

Overall 83/100 · Value 69Full breakdown →

What 100 OpenClaw requests cost on each pick

A typical request sends 99,534 input tokens, 87% of them read from cache, and a median session runs 16 requests. Measured on pooled coding-agent requests through our gateway (79,301 requests from 75 users; OpenClaw alone has fewer than 50 users, so its traffic is pooled), Sep 13 to Oct 12, 2026, priced at each vendor's list rates.

Model100 requestsMedian sessionMonth of regular use
Claude Sonnet 5.5$5.33$0.74$77.42
Claude Opus 5.5$10.66$1.47$155
GPT-6.1 Sol$6.74$0.74$90.04
Claude Haiku 5.5$1.31$0.096$17.73
DeepSeek V4.1 Flash$0.67$0.091$9.13
GLM 5.3$5.72$0.71$92.16

Month of regular use = the median user active on 5 or more days (1,478 requests in 30 days). Methodology · All models ranked by cost per request

Best model for OpenClaw under $19, $49 or $99 a month

The highest-rated model whose list-price cost for a month of regular use (1,478 requests) stays under each budget. Heavier months cost proportionally more.

Monthly budgetBest model that fitsEst. month
Under $19Claude Haiku 5.5$17.73
Under $49Claude Haiku 5.5$17.73
Under $99Claude Sonnet 5.5$77.42

"Highest-rated" uses our editorial quality and agentic scores; candidates are current models with a checked vendor price.

FAQ

What is the best model for OpenClaw?
Claude Sonnet 5.5 as the primary: 61.8% on Terminal-Bench 4.0 at $2/$10 per million tokens. Claude Opus 5.5 (64.9%) is worth it for long automations, and a model from another provider, such as GPT-6.1 Sol, makes a good first fallback.
How do OpenClaw fallbacks work?
OpenClaw tries the primary model first, rotates through that provider's auth profiles, and then moves down the fallback list in order. Its docs suggest fallbacks for cost- or latency-sensitive tasks and lower-stakes chat.
Which model keeps OpenClaw cheap to run?
Claude Haiku 5.5 and DeepSeek V4.1 Flash cost a fraction of Sonnet 5.5 per token. OpenClaw's docs warn against older or weaker models for tool-enabled agents and untrusted inputs, so use cheap models for chat and routine tasks and keep a strong primary.
Should I still use Claude Sonnet 4.6 or GPT-5.4 in OpenClaw?
Both have newer replacements at lower list prices: Claude Sonnet 5.5 costs $2/$10 per million tokens against Sonnet 4.6's $3/$15, and GPT-6.1 Sol $2/$10 against GPT-5.4's $2.50/$15.
How much do 100 OpenClaw requests cost?
At vendor list prices, 100 average coding-agent requests cost: Claude Sonnet 5.5 $5.33, Claude Opus 5.5 $10.66, GPT-6.1 Sol $6.74, Claude Haiku 5.5 $1.31, DeepSeek V4.1 Flash $0.67, GLM 5.3 $5.72. Measured on real coding-agent traffic through the Standard Compute gateway (Sep 13 to Oct 12, 2026), priced with each request's own token counts and cache reads.
What is the best model for OpenClaw on a $19, $49 or $99 monthly budget?
For a month of regular use (1,478 requests, the median in our data), the highest-rated model that fits under $19: Claude Haiku 5.5 (about $17.73); under $49: Claude Haiku 5.5 (about $17.73); under $99: Claude Sonnet 5.5 (about $77.42). Heavier months cost proportionally more.

The agent matters as much as the model. Compare AI agents →

Every model on this page runs on Standard Compute

Standard Compute

Current frontier models Claude Fable 5.1 · GPT-6 Astra

More intelligence
per dollar.

Each step goes to the cheapest model that can do it, on providers we vet for price, uptime and speed, with automatic fallback when one fails.

Compliance
Opt-in rules keep a company, team or key on European or US providers
Privacy
We don’t train on your prompts or store them

Fixed monthly plans from $19 · cancel anytime

7-day fair refund Free trial, no card Billing by Stripe4,100+ agent users