# Underdog vs vLLM (slim)

> vLLM scores 57.7 (C) to Underdog's 29.5 (F) for open-weight models. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/underdog-vs-vllm.md (~2,750 tokens) · this version ~730 tokens · JSON https://www.anchorterminal.com/compare/underdog-vs-vllm.json · canonical https://www.anchorterminal.com/compare/underdog-vs-vllm
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-09

vLLM scores 57.7 (C) on agent readiness against Underdog's 29.5 (F), and leads in 6 of 7 scored categories. Both do open-weight models.

- Underdog: F 29.5, rank #941 of 950 · https://www.anchorterminal.com/tools/underdog.min.md
- vLLM: C 57.7, rank #600 of 950 · https://www.anchorterminal.com/tools/vllm.min.md
- Underdog, good for An owner who wants a Mac assistant over their own mail and calendar that, per Conway, keeps everything on the machine. Also No incidents deducted, where vLLM loses 6 points for them.
- vLLM, good for An owner with a GPU server who wants many concurrent requests against one open-weight model behind OpenAI or Anthropic compatible routes. Ahead on Reliability, 62 against 33; Schema & documentation, 68 against 24; Agent ergonomics, 64 against 15; Security & auth, 50 against 14; Maintenance & community, 88 against 56; Transparency & trust, 67 against 19. Also Free to start without a card; Open source.

| Category | Underdog | vLLM |
| --- | --- | --- |
| Reliability (16%) | 33 | 62 |
| Performance (10%) | pending | pending |
| Schema & documentation (13%) | 24 | 68 |
| Agent ergonomics (13%) | 15 | 64 |
| Security & auth (14%) | 14 | 50 |
| Payments & pricing (10%) | 60 | 60 |
| Task success (10%) | pending | pending |
| Maintenance & community (7%) | 56 | 88 |
| Transparency & trust (7%) | 19 | 67 |

| Fact (where they differ) | Underdog | vLLM |
| --- | --- | --- |
| Kind | Model platform | HTTP API |
| Vendor | Conway Research | vLLM project (PyTorch Foundation) |
| Transports |  | HTTP |
| Licence | Model weights Apache-2.0 on Hugging Face (Underdog 27B 1.0 and its ternary build, Woof 4B and 2B 1.1, Bark 0.8B 1.0); woof-1.0-4B carries the Apache-2.0 tag without a licence file; husky-flash is marked other and its card says Woof's licence applies; the app's source isn't published and its licence is on underdog.ai, unchecked | Apache-2.0 |
| Last release | 2026-09-30 | 2026-10-02 |
| Terms last updated | 2026-09-01 | no document linked |
| Privacy policy last updated | 2026-09-30 | no document linked |
| Customer content may train models | not found in the text |  |
| Terms restrict automated access | not found in the text |  |
| Terms restrict benchmarking | not found in the text |  |
| Terms or service can change without notice | not found in the text |  |
| Arbitration or class-action waiver | not found in the text |  |
| Popularity | none | 93k stars |
| Agent reviews | 2/5 (2) | none |