beam
beam
Discover
PulseActivityAnalyticsBest forMapOrgs
Niches
AgentsMCPRAGCoding AssistantsInference & ServingVector DBs
Personal
WatchlistCompareWeekly report
?
Sign in
magic link · no password
LIVE──────── · ──:──:── UTCabout
beam
beam
Discover
PulseActivityAnalyticsBest forMapOrgs
Niches
AgentsMCPRAGCoding AssistantsInference & ServingVector DBs
Personal
WatchlistCompareWeekly report
?
Sign in
magic link · no password
Back to Pulse
Tool profile

youssofal/MTPLX

new

The fastest way to run Qwen 3.8 Flash Next, Qwen 3.8 27B and Ternary Bonsai 2 27B on a Mac: 125 tok/s in OpenCode on an M5 Max, and a 27B model on 16 GB Macs. Native MTP speculative decoding on Apple Silicon, exact at any temperature. OpenAI and Anthropic compatible local server.

anthropic-compatibleapple-siliconclaude-codeflash-nextinference-enginellm-inferencelocal-ailocal-llm
Velocity score
0.00/ 10
[STARS]
2.4k
[FORKS]
182
[CONTRIBUTORS]
44
[LAST_COMMIT]
10d ago
OPEN_ON_GITHUB
Velocity class: new
Is youssofal/MTPLX still actively maintained?
[VR]

This tool is in the Velocity Report when it moves. Get the weekly numbers.

Score breakdown
338/ 1000
coding-assistants · youssofal/MTPLX
Velocity50%
Adoption30%
Maintenance15%
Community5%
[CODE_GROWTH]
119
[INSTALL_VEL]
500
[ACTIVITY]
536
[COMMUNITY_SIGNAL]
968

Terminal score: 0–1000 raw, weighted across 4 dimensions. Public score: 0–10 normalized (shown in the 30-day stars chart above).

LIVE──────── · ──:──:── UTCabout