agentspeed.

together.ai

scanned 7/25/2026, 9:40:09 PM · cached · refresh in 21h · rubric 2026.07.4
Built with Webflow
robots.txt present: passrobots.txt allows AI agents: passContent-Signal directives: failllms.txt present: passSitemap present: passLink response headers: warnCanonical URL: passMCP server card: failOAuth authorization metadata: failOAuth resource metadata: failAPI catalog / OpenAPI: failWeb Bot Auth: failMarkdown negotiation: passText-to-markup ratio: failContent without JavaScript: skipHeading hierarchy: passCookie wall blocks content: passPage title: passMeta description: passJSON-LD present: passStructured data validates: warnSchema type coverage: failPaywall / login wall: passAgent Skills manifest: failWebMCP actions: failPrimary action reachable: passReachability: passTime to first byte: passFull render time: skipPage weight: pass80/ 100 · B
Good

This is how AI agents (not browsers) experience together.ai. The score weights five categories of machine-readability; the ticks on the arc are the 30 individual checks behind it.

✉ Email me this report + alert me when it changes ↓

Top fixes

Checks tagged “emerging” are 2026 agent-protocol standards most of the web hasn’t adopted yet — adopting early is an edge, not a defect.

failText-to-markup ratioReadability

Visible text is 3.3% of HTML weight (22283/668873 bytes). Low. Markup may be crowding out agent-usable text.

Fix: Heavy ad or navigation markup crowds out agent-usable text. Aim for a text-to-DOM ratio above 15%.

warnStructured data validatesStructured data

1 recommended-field warning(s). First: WebSite recommends "potentialAction" (improves agent comprehension).

Fix: Validate your JSON-LD with Google Rich Results test. Fix missing required properties.

failSchema type coverageStructured data

Page looks like FAQPage; missing JSON-LD type(s): FAQPage.

Fix: Align JSON-LD type to page purpose. Product pages without Product markup are skipped by agent shopping flows.

warnLink response headersDiscoverability

A Link: header is present but carries none of canonical/alternate/describedby — the relations agents consume.

Fix: Emit `Link:` headers for canonical, alternate-language, and describedby relations. Agents that fetch HEAD-only or stream-render rely on these.

failContent-Signal directivesemergingDiscoverability

No Content-Signal directives. Add e.g. `Content-Signal: ai-train=no, ai-summarize=yes` to declare granular AI usage policy beyond binary allow/disallow.

Fix: Add `Content-Signal:` directives to your robots.txt. This is the emerging standard (Cloudflare-driven) for declaring fine-grained AI usage policy beyond a binary allow/disallow.

failMCP server cardemergingDiscoverability

No valid MCP Server Card at /.well-known/mcp/server-card.json. Publish one so agents can discover your tools without HTML scraping.

Fix: Publish an MCP Server Card describing the tools your site exposes. Agents discover capabilities via `/.well-known/mcp/server-card.json`. See modelcontextprotocol.io for the schema.

Don’t lose this report

Get it in your inbox now — and a heads-up whenever together.ai’s agent-readiness score changes. Free, unsubscribe anytime.

Qualifies for certificationSilver

This score clears the Silver threshold (80+). Certification turns it into a dated, publicly verifiable attestation — a verification URL, an embeddable badge, and weekly monitoring for a full year. Scores under 70 can’t buy this, which is what makes displaying it mean something.

Get certified — $199/year →

What the agent receives

The agent’s-eye view
Together AI | The AI Native Cloud 💰 Announcing our Series C. Intelligence should be abundant, not expensive → 🤝 Together AI & Y Combinator announce partnership to deliver the first dedicated YC GPU cluster → ⚡ On-demand B200s now available on Together GPU Clusters → 🚀 Now serving MiniMax-M3 for efficient inference → Inference Serverless Inference High-performance inference as APIs Batch Inference Inference for batch workloads Provisioned Throughput Token-based capacity with SLAs Dedicated Model Inference Inference on custom hardware Dedicated Container Inference Inference for custom models MiniMax M3 Gemma 4 31B DeepSeek V4 Pro GLM-5.2 kimi K2.7 Code gpt-oss-120B Model library Explore the top open-source models Compute Accelerated Compute GPU Clusters Reliable GPU clusters at scale AI Factory Custom infrastructure at frontier scale Developer Environments Sandbox Build development environments for AI Storage Managed Storage Store model weights & data securely GB300 GB200 B200 H200 H100 Model Shaping Fine-Tuning Shape models with your data Evaluations Measure model quality kimi K2.7 Code Gemma 4 31B-it FP8 GLM 5.1 FP4 gpt-oss-120b Qwen3.5 397B A17b Llama 4 Maverick Model library Fine-tune top open-source models Research Research Systems research for production AI Research blog All our research publications Featured publications FlashAttention ATLAS Kernel Collection ThunderKittens DSGym Show all Developers Documentation Technical docs for Together AI Demos Our open-source demo apps Cookbooks Practical implementation guides Voice Agents Build voice agents for production Open-source AI Build better with open models Model Library Playground Together Chat Which LLM to use Open-source ROI calculator Company Resources Customer stories Testimonials from AI Natives Startup accelerator Build and scale your startup Customer support Find answers to your questions Blog Our latest news & blog posts Events Explore our events calendar Company About Get to know us Careers Join our mission Press Together in the news Pricing Serverless Inference High-performance inference as APIs Batch Inference Inference for batch workloads Provisioned Throughput Token-based capacity with SLAs Dedicated Model Inference Inference on custom hardware Dedicated Container Inference Inference for custom models MiniMax M3 Gemma 4 31B DeepSeek V4 Pro GLM-5.2 kimi K2.7 Code gpt-oss-120B Model library Explore the top open-source models Accelerated Compute GPU Clusters Reliable GPU clusters at scale AI Factory Custom infrastructure at frontier scale Developer Environments Sandbox Build development environments for AI Storage Managed Storage Store model weights & data securely GB300 GB200 B200 H200 H100 Fine-Tuning Shape models with your data Evaluations Measure model quality kimi K2.7 Code Gemma 4 31B-it FP8 GLM 5.1 FP4 gpt-oss-120b Qwen3.5 397B A17b Llama 4 Maverick Model library Fine-tune top open-source models Research Systems research for production AI Research blog All our research publications Featured publications FlashAttention ATLAS Kernel Collection ThunderKittens DSGym Show all Documentation Technical docs for Together AI Demos Our open-source demo apps Cookbooks Practical implementation guides Voice Agents Build voice agents for production Open-source AI Build better with open models Model Library Playground Together Chat Which LLM to use Open-source ROI calculator Resources Customer stories Testimonials from AI Natives Startup accelerator Build and scale your startup Customer support Find answers to your questions Blog Our latest news & blog posts Events Explore our events calendar Company About Get to know us Careers Join our mission Press Together in the news Contact sales Contact sales Sign in Build what's next on the AI Native Cloud Full-stack AI platform, powered by cutting-edge research. Start building Contact Sales Trusted by The Together AI Platform Accelerate inference, model shaping and pre-training on a research-optimized platform. Faster inference 2x powered by cutting-edge research. Learn how Lower cost 60% with workload-specific optimization. Learn how Faster pre-training 90% with Together Kernel Collection. Learn how Full-stack cloud Powering every step of the AI development journey —from experimentation to massive scale. Inference Compute Model shaping Serverless Inference The fastest way to run open-source models on demand. Powered by cutting-edge inference research. No infrastructure to manage, no long-term commitments. Learn more Batch Inference Cost-effectively process massive workloads asynchronously. Scale to 30 billion tokens per model with any serverless model or private deployment. Learn more Provisioned Throughput Committed inference capacity with token-based pricing, reserved throughput, and a 99% uptime SLA. Drop-in API compatibility for production workloads with no infrastructure to manage. Learn more Dedicated Model Inference Deploy models on dedicated infrastructure. Purpose-built for teams who need speed, control, and the best economics in the market. Learn more Dedicated Container Inference GPU infrastructure purpose-built for generative media workloads. Deploy video, audio, and image models with performance acceleration powered by Together Research. Learn more Accelerated Compute Scale from self-serve instant clusters to thousands of GPUs, all optimized for better performance with Together Kernel Collection. Learn more Sandbox Use fast, secure code sandboxes at scale to set up full-scale development environments for AI apps and agents. Learn more Managed Storage High-performance managed storage for AI-native workloads. Object storage and parallel filesystems optimized for AI, with zero egress fees. Learn more import { CodeSandbox } from "@codesandbox/sdk" const sdk = new CodeSandbox() const sandbox = await sdk.sandboxes.create({ id: "node-template" }) const client = await sandbox.connect() await client.commands.run("npm install && npm run build") const previewUrl = client.hosts.getUrl(3000) await sdk.sandboxes.hibernate(sandbox.id) // Snapshot saved — resume anytime in Fine-Tuning Fine-tune open-source models for production workloads, using the latest research techniques. Improve accuracy, reduce hallucinations, and control behavior — without managing training infrastructure. Learn more Grounded in cutting-edge research Foundational systems research for production AI. Together AI at ICML 2026: frontier research across the full stack Together Research Read More Kernels ParallelKernelBench: Frontier LLMs can't write fast multi-GPU kernels (yet) Willy Chan, Nathan Paek, Simon Guo, Simran Arora, Daniel Y. Fu Read More Agents Violin: An open-source video translation skill that breaks language barriers Shang Zhu, Kevin Qinghong Lin (Oxford), James Zou Read More Inference Accelerate RL rollouts by up to 50% with distribution-aware speculative decoding Zelei Shao, Vikranth Srivatsa, Sanjana Srivastava, Qingyang Wu, Alpay Ariyak, Xiaoxia Wu, Ameen Patel, Jue Wang, Percy Liang, Tri Dao, Ce Zhang, Yiying Zhang, Ben Athiwaratkun, Chenfeng Xu, Junxiong Wang Read More Architecture Parcae: Doing more with fewer parameters using stable looped models Hayden Prairie, Zachary Novack, Taylor Berg-Kirkpatrick, Dan Fu Read More Agents EinsteinArena: Harnessing the collective intelligence of agents in the wild to advance science Federico Bianchi,* Yongchan Kwon,* James Zou Read More Agents AI for Systems: Using LLMs to Optimize Database Query Execution Mehmet Hamza Erol, Xiangpeng Hao, Federico Bianchi, Ciro Greco, Jacopo Tagliabue, James Zou Read More Kernels Inside the Together AI kernels team Will Van Eaton Read More Inference Aurora Junxiong Wang, Fengxiang Bie, Jisen Li, Zhongzhu Zhou, Zelei Shao, Yubo Wang, Yinghui Liu, Qingyang Wu, Avner May, Sri Yanamandra, Ce Zhang, Tri Dao, Percy Liang, Shuaiwen Leon Song, Ben Athiwaratkun, Chenfeng Xu, Xiaoxia Wu Read More Agents Plan, divide, and conquer: How weak models excel at long context 

Show your score

Embed this badge on your site. It links back to this live scan and updates on every rescan.

AgentSpeed score for together.ai
<a href="https://agentspeed.com/scan/together.ai"><img src="https://agentspeed.com/badge/together.ai" alt="AgentSpeed agent-readiness score" height="56"></a>

Track this score

Get this report by email, then a heads-up when together.ai’s agent-readiness actually changes — a check regressing, or the composite dropping.

Two things never trigger an email: a score change caused by us publishing a new rubric, because that is a different instrument rather than a change to your site, and movement explained only by timing-derived checks, which shift run to run on a site nobody touched.

Want it on your own schedule, with Slack or a webhook instead of email? Score-drift monitoring is on every paid plan.

Full breakdown

16 pass2 warn10 fail2 skip
Discoverability
pass
robots.txt present
A robots.txt is reachable at the site root.
100/100
pass
robots.txt allows AI agents
All 8 answer-time access agents allowed. 1 training-data crawler(s) blocked (Google-Extended) — a licensing choice, not scored.
100/100
fail
Content-Signal directivesemerging
No Content-Signal directives. Add e.g. `Content-Signal: ai-train=no, ai-summarize=yes` to declare granular AI usage policy beyond binary allow/disallow.
0/100
pass
llms.txt present
Found /llms.txt (39088 bytes), H1 + at least one link present.
100/100
pass
Sitemap present
sitemap.xml reachable and referenced from robots.txt.
100/100
warn
Link response headers
A Link: header is present but carries none of canonical/alternate/describedby — the relations agents consume.
60/100
pass
Canonical URL
Canonical points to self: https://www.together.ai/.
100/100
fail
MCP server cardemerging
No valid MCP Server Card at /.well-known/mcp/server-card.json. Publish one so agents can discover your tools without HTML scraping.
0/100
fail
OAuth authorization metadataemerging
No RFC 8414 metadata at /.well-known/oauth-authorization-server. Agents that act on behalf of users need this to discover your authorization endpoints.
0/100
fail
OAuth resource metadataemerging
No RFC 9728 metadata at /.well-known/oauth-protected-resource. Publish it so agents can discover required scopes without hand-coded credentials.
0/100
fail
API catalog / OpenAPIemerging
No /.well-known/api-catalog (RFC 9727) and no /openapi.json|yaml. Publish one so agents that integrate with APIs can discover your endpoints.
0/100
fail
Web Bot Authemerging
No Web Bot Auth JWKS at /.well-known/http-message-signatures-directory.json. Publish one to allow trusted agents while keeping a default-deny posture for the rest.
0/100
Readability
pass
Markdown negotiationemerging
Requests with Accept: text/markdown return a markdown representation — agents prefer markdown over HTML for summarization and citation.
100/100
fail
Text-to-markup ratio
Visible text is 3.3% of HTML weight (22283/668873 bytes). Low. Markup may be crowding out agent-usable text.
25/100
skip
Content without JavaScript
Browser pass unavailable; cannot measure JS dependency.
pass
Heading hierarchy
Hierarchy is well-formed (1 H1, no skipped levels, 7 headings total).
100/100
pass
Cookie wall blocks content
No common consent-modal markers detected.
100/100
pass
Page title
A concise <title> is present (33 chars).
100/100
pass
Meta description
A well-sized meta description is present (145 chars).
100/100
Structured data
pass
JSON-LD present
Found 1 JSON-LD block.
100/100
warn
Structured data validates
1 recommended-field warning(s). First: WebSite recommends "potentialAction" (improves agent comprehension).
95/100
fail
Schema type coverage
Page looks like FAQPage; missing JSON-LD type(s): FAQPage.
0/100
Actionability
pass
Paywall / login wall
No paywall or login-wall detected on the landing URL.
100/100
fail
Agent Skills manifestemerging
No Agent Skills manifest at /.well-known/agent-skills.json. Enumerate the tasks agents can perform (search, add-to-cart, contact-support) so they pick the right one without scraping.
0/100
fail
WebMCP actionsemerging
No WebMCP detected. On pages with first-class actions (cart, support, account), embed a WebMCP server so on-page agents invoke tools directly.
0/100
pass
Primary action reachable
Primary offering identified ("AI Native Cloud platform for inference, compute, and model d") with price/CTA ("Contact Sales").
100/100
pass
Reachability
HTTP 200 after 1 redirect.
100/100
Performance
pass
Time to first byte
TTFB 133ms. Healthy.
100/100
skip
Full render time
Full-render time unavailable (browser pass skipped or failed).
pass
Page weight
Initial document is a lean 653 KB.
100/100
Beyond the scan

Running AI agents of your own? AgentSpeed also monitors them in production — every run, its latency, cost, and failures, with alerts when something breaks. Free for 10,000 events a month, no credit card.

Start watching free →See plans