I wanted a local A/B of Ornith 1.5 against “GLM 5.3 with good context.” GLM-5.3 is the same 753B base as 5.2 with extra post-training, and Z.ai has not published weights yet. So the honest local test is Ornith 1.5 Q6_K versus GLM-5.2 at a quant that still leaves KV headroom on 512 GB.
The 435 GB UD-Q4_K_XL we already had on disk is the quality quant, not the context quant. June work on this machine already got ugly past 32k with it. The pick for this A/B was Unsloth UD-IQ3_S (~288 GB on disk, eight shards) so the two residents sit in the same RAM envelope.
Sequential, one giant at a time, stock Homebrew llama-server b9700, thinking off, OpenAI tools. Download of IQ3_S ran while Ornith stayed up. Then Ornith baseline, then Ornith down, then GLM on a temp port at 64k context and q8 KV.
| Ornith 1.5 Q6_K | GLM-5.2 UD-IQ3_S | |
|---|---|---|
| Weights | official GGUF, 303 GiB file | Unsloth dynamic, 288 GB dir |
| Params (server) | 396.3B MoE | 753.9B MoE |
| RSS after load | ~310 GiB at 262k ctx | ~282 GiB at 64k ctx |
| Train window | 262,144 | 1,048,576 |
| Serve window this run | 262,144 | 65,536 |
| Exact string / add() / tools | PASS | PASS |
2-turn run_python loop | PASS (5.63 s + 0.52 s) | PASS (4.36 s + 0.53 s) |
| Decode (server eval) | 31.9–34.8 tok/s | 16.8–18.1 tok/s |
| 32k-class prefill | 36,029 toks, 100.1 s, PREFILL_OK | cancelled at 24,066 toks / 400 s |
Decode rows are llama-server slot eval timings, not Terminal-Bench accounting. Ornith 32k prompt was the same fox-sentence dump used on GLM. GLM prefill was still running (~60–80 tok/s and falling) when the HTTP client cancelled; the server stayed up.
Both models are real. Both emit OpenAI tool_calls[]. Both stopped after the fake tool result 2870 instead of looping. That kills the “one of them is unusable garbage” take from a two-turn smoke. It does not pick a winner on SWE.
The Ultra job is interactive local text with long prompts. Decode and prefill are the scarce resources. Ornith is about 2× on decode. Ornith finished a 36k prefill in 100 seconds. GLM had not finished 24k in 400 seconds. The 1M train window on the GLM card is not free on llama.cpp at this quant.
I would not put GLM IQ3_S on the default Hermes route. I would not displace DS4-Flash on the Sparks for either of these. Ornith stays the M3 Ultra candidate if we want a whole-box open 397B. GLM IQ3_S is a slow-lane spare if we later care about the 753B post-train, not the daily driver.
Related: Ornith 1.5 Q6_K 262k recipe, Ornith 1.0 GGUF + MLX, GLM-5.2 MXFP4 MLX recipe (different engine), GLM-5.2 Terminal-Bench.
All A/B numbers above came from direct HTTP calls on August 19–20, 2026 against localhost llama-server b9700 on the M3 Ultra. Official Ornith and Z.ai card scores are not reproduced here.