Halogen (@halogenengine) header
Halogen (@halogenengine) profile picture

halogen™ is local inference that behaves like a frontier API. Strix Halo + NPU, OpenAI & Anthropic APIs on one port.

Florida, USA
Joined October 2026
Photo by Halogen (@halogenengine)Photo by Halogen (@halogenengine)Photo by Halogen (@halogenengine)Photo by Halogen (@halogenengine)
Halogen 0.20.6 for Ryzen AI Halo: • Images read ~4x faster • New draft head: ~4% faster generation • Opt-in concurrency mode: up to 20% more throughput • @liquidai's own 4-bit LFM3-1.7B on the NPU, ~48 tok/s • Multi-agent & shared-prompt fixes github.com/peonist-ai/haloge…
1
5
1
19
715
Halogen 0.20.5 for Ryzen AI Halo: • Faster decoding with parallel requests • Hindi, Thai & Indic prompts read as trained • LFM3 from @liquidai on the @AIatAMD NPU: ~2x faster on long prompts • Fixes incl. structured output & Claude Code tool changes github.com/peonist-ai/haloge…
1
1
1
31
941
Halogen 0.20.4 is out for Strix Halo • Short follow-up messages read 15–50% faster in cached conversations • ~10–20% more combined throughput at 16 requests at once (>8 slots) • Fix: tool_choice “none” respected after earlier tool calls github.com/peonist-ai/haloge…
2
3
2
27
936
Halogen 0.20.2 and 0.20.3 are out for Strix Halo • @vllm_project’s d3-lite 14–18% faster on the @AIatAMD NPU, d3-nano 7–9%, both using less energy • Up to ~4% sooner first token on long conversations • NPU decisions on long prompts start sooner github.com/peonist-ai/haloge…
4
3
1
32
990
Halogen 0.20.0 is out for Strix Halo • OvisOCR3 document parsing on the @AIatAMD NPU (your own weights) • OCRBench 85.7 vs 85.3 upstream • ROCm 10.1: ~9–15% less energy, same speed • 2–6% faster prompt reading on Qwen3.8-Flash-Next github.com/peonist-ai/haloge…
4
9
2
67
4,190
Halogen 0.19.1 for Strix Halo: • XRT now in the image: NPU needs only the host driver and firmware • Malformed tool schemas get a 400 instead of locking up generation slots • Faster first token when a /v1/completions prompt extends a recent one github.com/peonist-ai/haloge…
2
7
252
Halogen 0.19.0 for @AIatAMD Strix Halo: • Preview: Halogen Spotter, an opt-in NPU agent guard that checks tool results • Whole-file rewrites by coding agents ~30% faster • NPU stability fixes Spotter: peonist.ai/halogen/docs/agen… Changelog: github.com/peonist-ai/haloge…
2
2
11
380
Halogen 0.18.23: the LFM3 family from @liquidai now runs on the Strix Halo NPU • LFM3-VL-2B: pictures and video • LFM3-Audio-2B: speech in and speech out • LFM3-Omni-2B: every medium on one model • Fix: replies no longer invent a user turn github.com/peonist-ai/haloge…
1
4
115
Another excellent video from @dcapitella Thanks for including Halogen! Strix Halo vs. two Intel Arc Pro B70s on Qwen3.8-Flash-Next: • ~64 tok/s at 64K context • 1,217 tok/s prompt processing • Close in generation, faster at prompt processing youtu.be/Rxh8ouzTEbM?t=1123
1
2
63
Wendell’s @Level1Techs review of the new @gmktecdotcom EVO-X5 Pro: with Halogen, Qwen3.8-Flash-Next on the Ryzen AI Max+ 495 runs at parity with DGX Spark. Thanks Wendell! youtu.be/YbIX2AWCqB8?t=621
2
4
229
Run your own @liquidai d1-2B on the @AIatAMD Strix Halo NPU. Halogen 0.18.22 is out. • Up to 16 questions on text, images or video in one pass • Finds objects in photos with bounding boxes • ~0.35 s per image question github.com/peonist-ai/haloge…
1
2
55
Run your own @liquidai LFM3-1.7B on the @AIatAMD Strix Halo NPU. Halogen 0.18.21 is out. Your checkpoint or fine-tune, up to ~38 tok/s Full 262k context Long NPU requests no longer stall behind a reply github.com/peonist-ai/haloge…
1
2
1
6
141
Halogen 0.18.18 is out. • Claude Code sessions with MCP tools keep their prompt cache • ggml-org's official IQ4_NL GGUF now loads • Short requests start during long prompts • Tool-call args with </tool_call> arrive whole Same outputs, same speed. github.com/peonist-ai/haloge…
1
1
5
61
Halogen 0.18.17 is out. Agents sharing a prompt no longer swap turns on a full KV pool Claude Code MCP tool search works on /v1/messages Tool-call args with markup arrive intact /v1/models reports image support Same outputs, same speed. github.com/peonist-ai/haloge…
2
2
38