Ternary Bonsai 2 9B runs at 77 tok/s (76.5 as recorded) on an M5 Pro, a published figure reported by github.com (llama.cpp, PQ2_0, tg128, batch 1). It needs about 5 GB of unified memory. It is Prism ML's 9.41B dense model, released under the Apache 2.0 licence.
Want to run Ternary Bonsai 2 9B faster — or step up to bigger models? Find your Mac →This is a published figure: a real run, reported by its source and linked below. The index records it as given and never adjusts it.
| Metric | Value |
|---|---|
| Tokens per second | sourced77 tok/s |
| Provenance | published figure · github.com |
| Quantization | PQ2_0 |
| Engine | llama.cpp |
| Context | tg128, batch 1 |
| Figure dated | 2026-10 |
| Minimum RAM | 5 GB |
| Parameters | 9.41B |
| Chip bandwidth | 307 GB/s |
Each row as the index holds it, with its source. “Mac RAM” is the memory of the Mac a run used, not the model’s requirement (5 GB). The headline figure follows the reference rule: the lower-median measured plain-decode run at 5 bits or fewer (a run whose quant was not stated counts only when no stated-quant run qualifies), else the estimate; a run marked “not used to anchor estimates” is listed here but never becomes the headline.
| Speed | Quant | Engine | Mac RAM | Context | Source | Note |
|---|---|---|---|---|---|---|
| sourced76.5 tok/s | PQ2_0 | llama.cpp | — | tg128, batch 1 | github.com | Vendor whitepaper Table 3: MacBook Pro M5 Pro, PrismML llama.cpp fork (Metal), llama-bench, flash attention on, mean of 5; 2.4 GB. |
Ollama — install it (brew install ollama or the app from ollama.com), then pull the model by its tag, checked on ollama.com on 2026-10-17:
MLX — pick a build from Hugging Face’s MLX listings, then run it with its repo id:
Downloads: Ollama page → · model card →
The 77 tok/s figure above came from llama.cpp at PQ2_0; other runtimes and quants run at different speeds.
Prefer a GUI? LM Studio downloads GGUF and MLX builds in one click. Step by step: install Ollama on a Mac.
Prism ML's ternary Qwen3.5-9B with vision (15 Oct 2026, Apache 2.0): 2.4 GB plus a 0.6 GB vision projector, 262K context. Ternary decoding is kernel-bound, so the index quotes only vendor figures: 76.5 tok/s on an M5 Pro (PrismML llama.cpp fork, PQ2_0, tg128). Ollama's bonsai-2:9b (3.2 GB) uses the Q2_0 pack.
Apache 2.0 licence · 9.41B · 262K context · LLMCheck capability 16/50 · see its leaderboard rank →
Ternary Bonsai 2 9B needs about 5 GB of unified memory. These current Apple Silicon Macs have the headroom to run it comfortably:
As an Amazon Associate, LLMCheck earns from qualifying purchases. These affiliate links cost you nothing extra and help keep the LLMCheck index free.
See how Ternary Bonsai 2 9B stacks up against other models on your specific Mac hardware.
Open Compare Tool Full Leaderboard