[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-together-ai-together-sandboxes":3,"mdc--gqf0w3-key":37,"related-repo-together-ai-together-sandboxes":516,"related-org-together-ai-together-sandboxes":624},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":32,"sourceUrl":35,"mdContent":36},"together-sandboxes","execute Python code in Together AI sandboxes","Remote Python execution in managed sandboxes on Together AI with stateful sessions, file uploads, data analysis, chart generation, and notebook-like runs via the Sandboxes API. Reach for it whenever the user wants managed remote Python execution instead of local execution, raw clusters, or full model hosting.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"together-ai","Together AI","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Ftogether-ai.jpg","togethercomputer",[13,17,20,23],{"name":14,"slug":15,"type":16},"Data Analysis","data-analysis","tag",{"name":18,"slug":19,"type":16},"Sandboxing","sandboxing",{"name":21,"slug":22,"type":16},"Python","python",{"name":24,"slug":25,"type":16},"Code Execution","code-execution",31,"https:\u002F\u002Fgithub.com\u002Ftogethercomputer\u002Fskills","2026-07-17T06:04:26.979066",null,4,[],{"repoUrl":27,"stars":26,"forks":30,"topics":33,"description":34},[],"Skills to help your coding agents use Together AI products.","https:\u002F\u002Fgithub.com\u002Ftogethercomputer\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Ftogether-sandboxes","---\nname: together-sandboxes\ndescription: \"Remote Python execution in managed sandboxes on Together AI with stateful sessions, file uploads, data analysis, chart generation, and notebook-like runs via the Sandboxes API. Reach for it whenever the user wants managed remote Python execution instead of local execution, raw clusters, or full model hosting.\"\n---\n\n# Together Sandboxes\n\n## Overview\n\nUse Together Sandboxes when the user wants to execute Python remotely in a managed sandbox.\n\nTypical fits:\n\n- stateful Python sessions\n- data analysis and chart generation\n- agent-generated code execution\n- file uploads into a remote runtime\n\n## When This Skill Wins\n\n- The user wants remote execution rather than local shell execution\n- Session state needs to persist across multiple calls\n- The result may include display outputs such as charts\n- A lightweight managed runtime is enough; no custom infra is required\n\n## Hand Off To Another Skill\n\n- Use `together-gpu-clusters` for full infrastructure control or larger distributed jobs\n- Use `together-dedicated-containers` for custom containerized runtime logic\n- Use `together-chat-completions` if the user only wants generated code, not executed code\n\n## Quick Routing\n\n- **Remote execution with session reuse**\n  - Start with [scripts\u002Fexecute_with_session.py](scripts\u002Fexecute_with_session.py) or [scripts\u002Fexecute_with_session.ts](scripts\u002Fexecute_with_session.ts)\n- **Response schema and session listing**\n  - Read [references\u002Fapi-reference.md](references\u002Fapi-reference.md)\n- **MCP-style access for agent workflows**\n  - Read [references\u002Fapi-reference.md](references\u002Fapi-reference.md)\n\n## Workflow\n\n1. Decide whether the task needs code execution or only code generation.\n2. Start a session with `client.code_interpreter.execute()`.\n3. Reuse `session_id` when the workflow depends on prior state.\n4. Inspect `stdout`, `stderr`, structured outputs, and display outputs separately.\n5. List sessions only when the user needs operational visibility or cleanup.\n\n## High-Signal Rules\n\n- Python scripts require the Together v2 SDK (`together>=2.0.0`). If the user is on an older version, they must upgrade first: `uv pip install --upgrade \"together>=2.0.0\"`.\n- Treat `session_id` as part of the workflow state.\n- Inspect `response.errors` before assuming a run succeeded.\n- `plt.show()` with the Agg backend does not reliably produce `display_data` outputs. To retrieve charts, save the figure to a `BytesIO` buffer with `fig.savefig()`, base64-encode it, and print the encoded string to stdout. Parse it from the `stdout` output on the client side. See the chart example in [scripts\u002Fexecute_with_session.py](scripts\u002Fexecute_with_session.py).\n- Use this skill when the user benefits from remote stateful execution, not just because Python is involved.\n- If the task outgrows the sandbox model, hand off to GPU clusters or dedicated containers.\n\n## Resource Map\n\n- **API reference**: [references\u002Fapi-reference.md](references\u002Fapi-reference.md)\n- **Alternative access patterns**: [references\u002Fapi-reference.md](references\u002Fapi-reference.md)\n- **Python workflow**: [scripts\u002Fexecute_with_session.py](scripts\u002Fexecute_with_session.py)\n- **TypeScript workflow**: [scripts\u002Fexecute_with_session.ts](scripts\u002Fexecute_with_session.ts)\n\n## Official Docs\n\n- [Together Sandboxes](https:\u002F\u002Fdocs.together.ai\u002Fdocs\u002Ftogether-code-interpreter)\n- [Sandboxes API](https:\u002F\u002Fdocs.together.ai\u002Freference\u002Ftci-execute)\n",{"data":38,"body":39},{"name":4,"description":6},{"type":40,"children":41},"root",[42,50,57,63,68,93,99,122,128,169,175,248,254,315,321,425,431,487,493],{"type":43,"tag":44,"props":45,"children":46},"element","h1",{"id":4},[47],{"type":48,"value":49},"text","Together Sandboxes",{"type":43,"tag":51,"props":52,"children":54},"h2",{"id":53},"overview",[55],{"type":48,"value":56},"Overview",{"type":43,"tag":58,"props":59,"children":60},"p",{},[61],{"type":48,"value":62},"Use Together Sandboxes when the user wants to execute Python remotely in a managed sandbox.",{"type":43,"tag":58,"props":64,"children":65},{},[66],{"type":48,"value":67},"Typical fits:",{"type":43,"tag":69,"props":70,"children":71},"ul",{},[72,78,83,88],{"type":43,"tag":73,"props":74,"children":75},"li",{},[76],{"type":48,"value":77},"stateful Python sessions",{"type":43,"tag":73,"props":79,"children":80},{},[81],{"type":48,"value":82},"data analysis and chart generation",{"type":43,"tag":73,"props":84,"children":85},{},[86],{"type":48,"value":87},"agent-generated code execution",{"type":43,"tag":73,"props":89,"children":90},{},[91],{"type":48,"value":92},"file uploads into a remote runtime",{"type":43,"tag":51,"props":94,"children":96},{"id":95},"when-this-skill-wins",[97],{"type":48,"value":98},"When This Skill Wins",{"type":43,"tag":69,"props":100,"children":101},{},[102,107,112,117],{"type":43,"tag":73,"props":103,"children":104},{},[105],{"type":48,"value":106},"The user wants remote execution rather than local shell execution",{"type":43,"tag":73,"props":108,"children":109},{},[110],{"type":48,"value":111},"Session state needs to persist across multiple calls",{"type":43,"tag":73,"props":113,"children":114},{},[115],{"type":48,"value":116},"The result may include display outputs such as charts",{"type":43,"tag":73,"props":118,"children":119},{},[120],{"type":48,"value":121},"A lightweight managed runtime is enough; no custom infra is required",{"type":43,"tag":51,"props":123,"children":125},{"id":124},"hand-off-to-another-skill",[126],{"type":48,"value":127},"Hand Off To Another Skill",{"type":43,"tag":69,"props":129,"children":130},{},[131,145,157],{"type":43,"tag":73,"props":132,"children":133},{},[134,136,143],{"type":48,"value":135},"Use ",{"type":43,"tag":137,"props":138,"children":140},"code",{"className":139},[],[141],{"type":48,"value":142},"together-gpu-clusters",{"type":48,"value":144}," for full infrastructure control or larger distributed jobs",{"type":43,"tag":73,"props":146,"children":147},{},[148,149,155],{"type":48,"value":135},{"type":43,"tag":137,"props":150,"children":152},{"className":151},[],[153],{"type":48,"value":154},"together-dedicated-containers",{"type":48,"value":156}," for custom containerized runtime logic",{"type":43,"tag":73,"props":158,"children":159},{},[160,161,167],{"type":48,"value":135},{"type":43,"tag":137,"props":162,"children":164},{"className":163},[],[165],{"type":48,"value":166},"together-chat-completions",{"type":48,"value":168}," if the user only wants generated code, not executed code",{"type":43,"tag":51,"props":170,"children":172},{"id":171},"quick-routing",[173],{"type":48,"value":174},"Quick Routing",{"type":43,"tag":69,"props":176,"children":177},{},[178,208,229],{"type":43,"tag":73,"props":179,"children":180},{},[181,187],{"type":43,"tag":182,"props":183,"children":184},"strong",{},[185],{"type":48,"value":186},"Remote execution with session reuse",{"type":43,"tag":69,"props":188,"children":189},{},[190],{"type":43,"tag":73,"props":191,"children":192},{},[193,195,201,203],{"type":48,"value":194},"Start with ",{"type":43,"tag":196,"props":197,"children":199},"a",{"href":198},"scripts\u002Fexecute_with_session.py",[200],{"type":48,"value":198},{"type":48,"value":202}," or ",{"type":43,"tag":196,"props":204,"children":206},{"href":205},"scripts\u002Fexecute_with_session.ts",[207],{"type":48,"value":205},{"type":43,"tag":73,"props":209,"children":210},{},[211,216],{"type":43,"tag":182,"props":212,"children":213},{},[214],{"type":48,"value":215},"Response schema and session listing",{"type":43,"tag":69,"props":217,"children":218},{},[219],{"type":43,"tag":73,"props":220,"children":221},{},[222,224],{"type":48,"value":223},"Read ",{"type":43,"tag":196,"props":225,"children":227},{"href":226},"references\u002Fapi-reference.md",[228],{"type":48,"value":226},{"type":43,"tag":73,"props":230,"children":231},{},[232,237],{"type":43,"tag":182,"props":233,"children":234},{},[235],{"type":48,"value":236},"MCP-style access for agent workflows",{"type":43,"tag":69,"props":238,"children":239},{},[240],{"type":43,"tag":73,"props":241,"children":242},{},[243,244],{"type":48,"value":223},{"type":43,"tag":196,"props":245,"children":246},{"href":226},[247],{"type":48,"value":226},{"type":43,"tag":51,"props":249,"children":251},{"id":250},"workflow",[252],{"type":48,"value":253},"Workflow",{"type":43,"tag":255,"props":256,"children":257},"ol",{},[258,263,276,289,310],{"type":43,"tag":73,"props":259,"children":260},{},[261],{"type":48,"value":262},"Decide whether the task needs code execution or only code generation.",{"type":43,"tag":73,"props":264,"children":265},{},[266,268,274],{"type":48,"value":267},"Start a session with ",{"type":43,"tag":137,"props":269,"children":271},{"className":270},[],[272],{"type":48,"value":273},"client.code_interpreter.execute()",{"type":48,"value":275},".",{"type":43,"tag":73,"props":277,"children":278},{},[279,281,287],{"type":48,"value":280},"Reuse ",{"type":43,"tag":137,"props":282,"children":284},{"className":283},[],[285],{"type":48,"value":286},"session_id",{"type":48,"value":288}," when the workflow depends on prior state.",{"type":43,"tag":73,"props":290,"children":291},{},[292,294,300,302,308],{"type":48,"value":293},"Inspect ",{"type":43,"tag":137,"props":295,"children":297},{"className":296},[],[298],{"type":48,"value":299},"stdout",{"type":48,"value":301},", ",{"type":43,"tag":137,"props":303,"children":305},{"className":304},[],[306],{"type":48,"value":307},"stderr",{"type":48,"value":309},", structured outputs, and display outputs separately.",{"type":43,"tag":73,"props":311,"children":312},{},[313],{"type":48,"value":314},"List sessions only when the user needs operational visibility or cleanup.",{"type":43,"tag":51,"props":316,"children":318},{"id":317},"high-signal-rules",[319],{"type":48,"value":320},"High-Signal Rules",{"type":43,"tag":69,"props":322,"children":323},{},[324,344,356,368,415,420],{"type":43,"tag":73,"props":325,"children":326},{},[327,329,335,337,343],{"type":48,"value":328},"Python scripts require the Together v2 SDK (",{"type":43,"tag":137,"props":330,"children":332},{"className":331},[],[333],{"type":48,"value":334},"together>=2.0.0",{"type":48,"value":336},"). If the user is on an older version, they must upgrade first: ",{"type":43,"tag":137,"props":338,"children":340},{"className":339},[],[341],{"type":48,"value":342},"uv pip install --upgrade \"together>=2.0.0\"",{"type":48,"value":275},{"type":43,"tag":73,"props":345,"children":346},{},[347,349,354],{"type":48,"value":348},"Treat ",{"type":43,"tag":137,"props":350,"children":352},{"className":351},[],[353],{"type":48,"value":286},{"type":48,"value":355}," as part of the workflow state.",{"type":43,"tag":73,"props":357,"children":358},{},[359,360,366],{"type":48,"value":293},{"type":43,"tag":137,"props":361,"children":363},{"className":362},[],[364],{"type":48,"value":365},"response.errors",{"type":48,"value":367}," before assuming a run succeeded.",{"type":43,"tag":73,"props":369,"children":370},{},[371,377,379,385,387,393,395,401,403,408,410,414],{"type":43,"tag":137,"props":372,"children":374},{"className":373},[],[375],{"type":48,"value":376},"plt.show()",{"type":48,"value":378}," with the Agg backend does not reliably produce ",{"type":43,"tag":137,"props":380,"children":382},{"className":381},[],[383],{"type":48,"value":384},"display_data",{"type":48,"value":386}," outputs. To retrieve charts, save the figure to a ",{"type":43,"tag":137,"props":388,"children":390},{"className":389},[],[391],{"type":48,"value":392},"BytesIO",{"type":48,"value":394}," buffer with ",{"type":43,"tag":137,"props":396,"children":398},{"className":397},[],[399],{"type":48,"value":400},"fig.savefig()",{"type":48,"value":402},", base64-encode it, and print the encoded string to stdout. Parse it from the ",{"type":43,"tag":137,"props":404,"children":406},{"className":405},[],[407],{"type":48,"value":299},{"type":48,"value":409}," output on the client side. See the chart example in ",{"type":43,"tag":196,"props":411,"children":412},{"href":198},[413],{"type":48,"value":198},{"type":48,"value":275},{"type":43,"tag":73,"props":416,"children":417},{},[418],{"type":48,"value":419},"Use this skill when the user benefits from remote stateful execution, not just because Python is involved.",{"type":43,"tag":73,"props":421,"children":422},{},[423],{"type":48,"value":424},"If the task outgrows the sandbox model, hand off to GPU clusters or dedicated containers.",{"type":43,"tag":51,"props":426,"children":428},{"id":427},"resource-map",[429],{"type":48,"value":430},"Resource Map",{"type":43,"tag":69,"props":432,"children":433},{},[434,448,461,474],{"type":43,"tag":73,"props":435,"children":436},{},[437,442,444],{"type":43,"tag":182,"props":438,"children":439},{},[440],{"type":48,"value":441},"API reference",{"type":48,"value":443},": ",{"type":43,"tag":196,"props":445,"children":446},{"href":226},[447],{"type":48,"value":226},{"type":43,"tag":73,"props":449,"children":450},{},[451,456,457],{"type":43,"tag":182,"props":452,"children":453},{},[454],{"type":48,"value":455},"Alternative access patterns",{"type":48,"value":443},{"type":43,"tag":196,"props":458,"children":459},{"href":226},[460],{"type":48,"value":226},{"type":43,"tag":73,"props":462,"children":463},{},[464,469,470],{"type":43,"tag":182,"props":465,"children":466},{},[467],{"type":48,"value":468},"Python workflow",{"type":48,"value":443},{"type":43,"tag":196,"props":471,"children":472},{"href":198},[473],{"type":48,"value":198},{"type":43,"tag":73,"props":475,"children":476},{},[477,482,483],{"type":43,"tag":182,"props":478,"children":479},{},[480],{"type":48,"value":481},"TypeScript workflow",{"type":48,"value":443},{"type":43,"tag":196,"props":484,"children":485},{"href":205},[486],{"type":48,"value":205},{"type":43,"tag":51,"props":488,"children":490},{"id":489},"official-docs",[491],{"type":48,"value":492},"Official Docs",{"type":43,"tag":69,"props":494,"children":495},{},[496,506],{"type":43,"tag":73,"props":497,"children":498},{},[499],{"type":43,"tag":196,"props":500,"children":504},{"href":501,"rel":502},"https:\u002F\u002Fdocs.together.ai\u002Fdocs\u002Ftogether-code-interpreter",[503],"nofollow",[505],{"type":48,"value":49},{"type":43,"tag":73,"props":507,"children":508},{},[509],{"type":43,"tag":196,"props":510,"children":513},{"href":511,"rel":512},"https:\u002F\u002Fdocs.together.ai\u002Freference\u002Ftci-execute",[503],[514],{"type":48,"value":515},"Sandboxes API",{"items":517,"total":623},[518,537,553,566,584,594,609],{"slug":519,"name":519,"fn":520,"description":521,"org":522,"tags":523,"stars":26,"repoUrl":27,"updatedAt":536},"together-audio","process audio with Together AI","Text-to-speech and speech-to-text via Together AI, including REST, streaming, and realtime WebSocket TTS, plus transcription, translation, diarization, timestamps, and live STT. Reach for it whenever the user needs audio in or audio out on Together AI rather than chat generation, image or video creation, or model training.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[524,527,530,533],{"name":525,"slug":526,"type":16},"Audio","audio",{"name":528,"slug":529,"type":16},"Speech","speech",{"name":531,"slug":532,"type":16},"Text-to-Speech","text-to-speech",{"name":534,"slug":535,"type":16},"Transcription","transcription","2026-07-26T05:49:08.246858",{"slug":538,"name":538,"fn":539,"description":540,"org":541,"tags":542,"stars":26,"repoUrl":27,"updatedAt":552},"together-batch-inference","run asynchronous batch inference jobs","High-volume, asynchronous offline inference at up to 50% lower cost via Together AI's Batch API. Prepare JSONL inputs, upload files, create jobs, poll status, and download outputs. Reach for it whenever the user needs non-interactive bulk inference rather than real-time chat or evaluation jobs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[543,546,549],{"name":544,"slug":545,"type":16},"Automation","automation",{"name":547,"slug":548,"type":16},"LLM","llm",{"name":550,"slug":551,"type":16},"Performance","performance","2026-07-17T06:08:23.919602",{"slug":166,"name":166,"fn":554,"description":555,"org":556,"tags":557,"stars":26,"repoUrl":27,"updatedAt":565},"generate text with Together AI","Real-time and streaming text generation via Together AI's OpenAI-compatible chat\u002Fcompletions API, including multi-turn conversations, tool and function calling, structured JSON outputs, and reasoning models. Reach for it whenever the user wants to build or debug text generation on Together AI, unless they specifically need batch jobs, embeddings, fine-tuning, dedicated endpoints, dedicated containers, or GPU clusters.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[558,561,564],{"name":559,"slug":560,"type":16},"AI","ai",{"name":562,"slug":563,"type":16},"API Development","api-development",{"name":547,"slug":548,"type":16},"2026-07-26T05:49:11.244104",{"slug":154,"name":154,"fn":567,"description":568,"org":569,"tags":570,"stars":26,"repoUrl":27,"updatedAt":583},"deploy custom inference containers on Together AI","Custom Dockerized inference workers on Together AI's managed GPU infrastructure. Build with Sprocket SDK, configure with Jig CLI, submit async queue jobs, and poll results. Reach for it whenever the user needs container-level control rather than a standard model endpoint or raw cluster.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[571,574,577,580],{"name":572,"slug":573,"type":16},"AI Infrastructure","ai-infrastructure",{"name":575,"slug":576,"type":16},"Deployment","deployment",{"name":578,"slug":579,"type":16},"Docker","docker",{"name":581,"slug":582,"type":16},"Machine Learning","machine-learning","2026-07-26T05:49:09.267892",{"slug":585,"name":585,"fn":586,"description":587,"org":588,"tags":589,"stars":26,"repoUrl":27,"updatedAt":593},"together-dedicated-model-inference","deploy and operate Together AI models","Deploy and operate models on dedicated GPUs with Together AI's Dedicated Model Inference (DMI, the v2 dedicated endpoints API): beta endpoints, deployments, deployment profiles and hardware configs, autoscaling, traffic splitting, A\u002FB tests, shadow experiments, Prometheus metrics, and custom model or LoRA adapter uploads. Reach for it whenever the user mentions together beta endpoints or tg beta commands, client.beta.endpoints, DMI resources like ep_\u002Fdep_\u002Fcr_\u002Fml_ IDs, or wants production model serving with traffic management on Together AI. This is the current dedicated-hosting API and also covers migrating off the retired legacy v1 endpoints API (non-beta client.endpoints \u002F together endpoints), whose create and restart now return HTTP 403.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[590,591,592],{"name":572,"slug":573,"type":16},{"name":575,"slug":576,"type":16},{"name":547,"slug":548,"type":16},"2026-07-26T06:08:44.044039",{"slug":595,"name":595,"fn":596,"description":597,"org":598,"tags":599,"stars":26,"repoUrl":27,"updatedAt":608},"together-embeddings","generate embeddings and build RAG pipelines","Dense vector embeddings, semantic search, RAG pipelines, and reranking via Together AI. Generate embeddings with open-source models and rerank results behind dedicated endpoints. Reach for it whenever the user needs vector representations or retrieval quality improvements rather than direct text generation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[600,601,602,605],{"name":559,"slug":560,"type":16},{"name":547,"slug":548,"type":16},{"name":603,"slug":604,"type":16},"RAG","rag",{"name":606,"slug":607,"type":16},"Search","search","2026-07-26T05:49:06.247906",{"slug":610,"name":610,"fn":611,"description":612,"org":613,"tags":614,"stars":26,"repoUrl":27,"updatedAt":622},"together-evaluations","evaluate LLM outputs with Together AI","LLM-as-a-judge evaluation framework on Together AI. Classify, score, and compare model outputs, select judge models, use external-provider judges or targets, poll results and download reports. Reach for it whenever the user wants to benchmark outputs, grade responses, compare A\u002FB variants, or operationalize automated evaluations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[615,618,621],{"name":616,"slug":617,"type":16},"Benchmarking","benchmarking",{"name":619,"slug":620,"type":16},"Evals","evals",{"name":547,"slug":548,"type":16},"2026-07-26T05:49:07.241553",14,{"items":625,"total":747},[626,647,659,666,672,678,685,691,698,704,714,728],{"slug":627,"name":627,"fn":628,"description":629,"org":630,"tags":631,"stars":644,"repoUrl":645,"updatedAt":646},"blog-post","write structured blog posts","Write structured long-form blog posts with a consistent structure and SEO optimization.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[632,635,638,641],{"name":633,"slug":634,"type":16},"Content Creation","content-creation",{"name":636,"slug":637,"type":16},"Marketing","marketing",{"name":639,"slug":640,"type":16},"SEO","seo",{"name":642,"slug":643,"type":16},"Writing","writing",1151,"https:\u002F\u002Fgithub.com\u002Ftogethercomputer\u002Ftogether-cookbook","2026-07-17T06:07:39.475433",{"slug":648,"name":648,"fn":649,"description":650,"org":651,"tags":652,"stars":644,"repoUrl":645,"updatedAt":658},"social-media","create social media content","Create social media content including Twitter\u002FX threads, LinkedIn posts, and short-form updates.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[653,654,655,657],{"name":633,"slug":634,"type":16},{"name":636,"slug":637,"type":16},{"name":656,"slug":648,"type":16},"Social Media",{"name":642,"slug":643,"type":16},"2026-07-17T06:07:38.455058",{"slug":519,"name":519,"fn":520,"description":521,"org":660,"tags":661,"stars":26,"repoUrl":27,"updatedAt":536},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[662,663,664,665],{"name":525,"slug":526,"type":16},{"name":528,"slug":529,"type":16},{"name":531,"slug":532,"type":16},{"name":534,"slug":535,"type":16},{"slug":538,"name":538,"fn":539,"description":540,"org":667,"tags":668,"stars":26,"repoUrl":27,"updatedAt":552},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[669,670,671],{"name":544,"slug":545,"type":16},{"name":547,"slug":548,"type":16},{"name":550,"slug":551,"type":16},{"slug":166,"name":166,"fn":554,"description":555,"org":673,"tags":674,"stars":26,"repoUrl":27,"updatedAt":565},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[675,676,677],{"name":559,"slug":560,"type":16},{"name":562,"slug":563,"type":16},{"name":547,"slug":548,"type":16},{"slug":154,"name":154,"fn":567,"description":568,"org":679,"tags":680,"stars":26,"repoUrl":27,"updatedAt":583},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[681,682,683,684],{"name":572,"slug":573,"type":16},{"name":575,"slug":576,"type":16},{"name":578,"slug":579,"type":16},{"name":581,"slug":582,"type":16},{"slug":585,"name":585,"fn":586,"description":587,"org":686,"tags":687,"stars":26,"repoUrl":27,"updatedAt":593},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[688,689,690],{"name":572,"slug":573,"type":16},{"name":575,"slug":576,"type":16},{"name":547,"slug":548,"type":16},{"slug":595,"name":595,"fn":596,"description":597,"org":692,"tags":693,"stars":26,"repoUrl":27,"updatedAt":608},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[694,695,696,697],{"name":559,"slug":560,"type":16},{"name":547,"slug":548,"type":16},{"name":603,"slug":604,"type":16},{"name":606,"slug":607,"type":16},{"slug":610,"name":610,"fn":611,"description":612,"org":699,"tags":700,"stars":26,"repoUrl":27,"updatedAt":622},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[701,702,703],{"name":616,"slug":617,"type":16},{"name":619,"slug":620,"type":16},{"name":547,"slug":548,"type":16},{"slug":705,"name":705,"fn":706,"description":707,"org":708,"tags":709,"stars":26,"repoUrl":27,"updatedAt":713},"together-fine-tuning","fine-tune and adapt models on Together AI","LoRA, full fine-tuning, DPO preference tuning, VLM training, function-calling tuning, reasoning tuning, and BYOM uploads on Together AI. Reach for it whenever the user wants to adapt a model on custom data rather than only run inference, evaluate outputs, or host an existing model.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[710,711,712],{"name":559,"slug":560,"type":16},{"name":547,"slug":548,"type":16},{"name":581,"slug":582,"type":16},"2026-07-26T05:49:10.243114",{"slug":142,"name":142,"fn":715,"description":716,"org":717,"tags":718,"stars":26,"repoUrl":27,"updatedAt":727},"orchestrate GPU clusters on Together AI","On-demand and reserved GPU clusters (H100, H200, B200) on Together AI with Kubernetes or Slurm orchestration, shared storage, credential management, and cluster scaling for ML and HPC jobs. Reach for it when the user needs multi-node compute or infrastructure control rather than a managed model endpoint.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[719,720,723,726],{"name":572,"slug":573,"type":16},{"name":721,"slug":722,"type":16},"Cloud","cloud",{"name":724,"slug":725,"type":16},"Kubernetes","kubernetes",{"name":581,"slug":582,"type":16},"2026-07-26T05:49:05.252646",{"slug":729,"name":729,"fn":730,"description":731,"org":732,"tags":733,"stars":26,"repoUrl":27,"updatedAt":746},"together-images","generate and edit images with Together AI","Text-to-image generation and image editing via Together AI, including FLUX and Kontext models, LoRA-based styling, reference-image guidance, and local image downloads. Reach for it whenever the user wants to generate or edit images on Together AI rather than create videos or build text-only chat applications.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[734,737,740,743],{"name":735,"slug":736,"type":16},"Creative","creative",{"name":738,"slug":739,"type":16},"Design","design",{"name":741,"slug":742,"type":16},"Image Generation","image-generation",{"name":744,"slug":745,"type":16},"Multimodal","multimodal","2026-07-17T06:04:23.067755",16]