Self-Hosted Agentic LLMs
As of August 2026, AIDiveForge tracks 4 self-hosted agentic llms. The top three by verified-data score are Ornith-1.0, Codeium, and Command R7B. Curated self-hosted agentic llms tracked by AIDiveForge. Listings are verified against each tool's live website and re-checked regularly.
Last updated July 29, 2026 · 4 tools
Ranked by AIDiveForge's verified-data score: data completeness, verification recency, community rating, and real visitor engagement. How we rank · No tool can pay for placement.
1. Ornith-1.0
The model family spans 9B-Dense through 397B-MoE, all post-trained on Gemma 4 and Qwen 3.5 with a reinforcement learning framework that jointly optimizes the agent scaffold and the solution rollouts it produces. The vendor states benchmark performance on Terminal-Bench 2.1, SWE-Bench, NL2Repo, and OpenClaw at the top of the open-source range for each size tier. There is no hosted API — you pull the weights and run inference yourself, which means your infra team owns the deployment stack from day one. The 397B-MoE variant requires hardware that most teams do not have on-hand, so realistic entry for self-hosted production starts at the 31B-Dense tier. Community activity is early-stage: ten commits in the repository, six open issues, no closed pull requests.
FreeOpen SourceSelf-hostedVerified Jun 30, 2026
2. Codeium
Devin, from Cognition, operates as a self-directed agent: given a task, it plans steps, writes and executes code, runs tests, interprets the output, and iterates — without a developer holding its hand through each transition. The vendor positions it for high-volume routine tickets, legacy migrations, and exploratory codebase work where the bottleneck is throughput, not creativity. Teams delegate backlog tickets and get draft PRs back; the agent handles the scaffolding. The ceiling appears on tasks requiring deep organizational context — tribal knowledge about why a module exists, or business logic that lives in nobody's head and in no doc. At that point, a developer re-enters the loop, which partly offsets the delegation gain.
Paid$20/moAPISelf-hostedVerified Jun 3, 2026
3. Command R7B
Command R7B is a smaller language model optimized for tasks that don't require reasoning at the frontier—summarization, classification, instruction-following, and document analysis. Cohere positions it as the pragmatic choice for teams tired of paying for (or waiting on) 70B+ parameter models when a tighter, faster alternative works. It's free and open source, which means no API charges and full control over deployment. The real limitation: it will struggle on abstract reasoning, mathematical proof, or multi-step logic puzzles where 70B models shine. For enterprises choosing between this and proprietary APIs, the tradeoff is real but worth calculating.
PaidOpen SourcePay-as-you-goAPISelf-hosted128k tokens context$0.04/1M inputVerified Apr 9, 2026
4. Mistral Large 2
Mistral Large 2 is a general-purpose language model trained to handle complex reasoning, code generation, and multilingual work at the scale enterprises need. It's free to use via API or self-host, sits in the same performance tier as proprietary models from OpenAI and Anthropic, and can ingest documents up to 128,000 tokens long. The core trade-off: it has a knowledge cutoff earlier than competitors and lacks serious vision capabilities, making it less suitable for tasks requiring current events or image understanding. For teams optimizing on cost and reasoning quality rather than breadth of modalities, it's a genuine alternative to paid tiers.
FreeOpen SourceAPISelf-hosted262.1k tokens context$0.5/1M inputVerified Apr 8, 2026
Listings on this page are sourced and verified by the AIDiveForge data pipeline. AIDiveForge is editorially independent — no money changes hands for inclusion.