Qwen 3.6 35B A3B GGUF

✓ Verified by Badgr · llama.cpp · Serve

Last verified 2026-10-09

badgr serve --runtime llama.cpp --hf-repo unsloth/Qwen3.6-35B-A3B-GGUF --hf-file Qwen3.6-35B-A3B-UD-IQ4_XS.gguf --gpu auto --max-cost 3 --startup-timeout 50

GPU

NVIDIA GeForce RTX 3090Reported by the provider, not confirmed inside the container

Time to verified

186.1s

Deployment

dep-03c936ee8b

• ✓ llama-server started
• ✓ /health returned HTTP 200
• ✓ Tool-calling requests returned parsed tool_calls
• ✓ final_state = VERIFIED
• ✓ Teardown completed

A 17.7 GB model: Badgr sizes the startup window to the file, so the download is not abandoned mid-way. In 24 requests with the payload from ggml-org/llama.cpp#30118, 7 returned parsed tool_calls and none leaked <function=...> text into the message content.

GitHub issue this run was for: ggml-org/llama.cpp#30118

final_state=VERIFIED failure_class=None providers_tried=1 spend_usd=0.0147 elapsed=186.1s

Run from Badgr's local development environment. Documents that this deployment path works end-to-end for this model.

✓ Verified by Badgr · Ollama · Serve

Last verified 2026-10-09

badgr serve qwen3.6:35b --runtime ollama --gpu auto --max-cost 2 --startup-timeout 40

GPU

NVIDIA A40Reported by the provider, not confirmed inside the container

Time to verified

158s

Deployment

dep-ab1f15ad47

• ✓ Ollama server started
• ✓ /api/tags listed qwen3.6:35b
• ✓ A 10,101-token /api/chat request with num_ctx 65536 returned a reply
• ✓ final_state = VERIFIED
• ✓ Teardown completed

This is the Ollama library tag qwen3.6:35b (22.6 GB), not the Unsloth GGUF file used by the llama.cpp entry. A request with a 65,536-token context and a 10,101-token prompt completed. The elapsed time is estimated from the spend at verification and the hourly rate.

GitHub issue this run was for: ollama/ollama#18856

final_state=VERIFIED failure_class=None providers_tried=1 spend_usd=0.02

Run from Badgr's local development environment. Documents that this deployment path works end-to-end for this model.