GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
lewisdog
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
lewisdog
Model Tree
Qwen/Qwen3-1.7B
Input Modalities
Output Modalities
Supported Functionality
sh
python3 -m mlx_lm.convert --hf-path lewisdog/qwen3-1.7b-cogs-ask -q --q-bits 8 --mlx-path qwen3-cogs-ask-mlx# NOTE: if transformers writes v5-format configs, backfill rope_theta / rope_scaling /# torch_dtype from the base Qwen3-1.7B config and swap in the original Qwen3-1.7B# tokenizer files before converting. Wire as [llm.ask] in cogs.toml.Decode at temp 0 + repetition_penalty=1.1 (pure greedy runs away on long
list outputs), enable_thinking=False, stop on <|im_end|>.
Repair citation prefixes. Most citation errors are a dropped/wrong
namespace prefix (enisa-… → sources-enisa-…). Normalize against the notes
actually in the prompt and drop anything unverifiable:
python
PFX = ("sources-", "source-", "concepts-", "concept-", "entities-", "entity-")core = lambda x: next((x[len(p):] for p in PFX if x.startswith(p)), x)def repair(cite, note_ids): if cite in note_ids: return cite return {core(n): n for n in note_ids}.get(core(cite)) # None => drop itAbstention is reliable — trust it to decline out-of-domain questions.
LoRA r=32/α=64, 5 epochs, eff. batch 16, max_seq 8192, lr 1e-4 cosine, bf16, on a
DGX Spark (GB10). Data: cogs distill ask pairs (475 base + abstentions
oversampled 3×). Eval loss 1.906 → 0.758; eval token-acc 0.836. Full write-up in
the repo RESULTS.md.
PFX = ("sources-", "source-", "concepts-", "concept-", "entities-", "entity-")core = lambda x: next((x[len(p):] for p in PFX if x.startswith(p)), x)def repair(cite, note_ids): if cite in note_ids: return cite return {core(n): n for n in note_ids}.get(core(cite)) # None => drop it