Issues

• hola

  microsoft/BitNet#645 · Filter by author a88808632-star opened on Oct 12, 2026

• macOS / Apple Silicon: setup_env.py build fails (ggml-base can't find i2_s symbols); after working around it, Metal crashes and CPU output is garbage

  microsoft/BitNet#641 · Filter by author AdamMiltonBarker opened on Oct 11, 2026
  1

• Published BitNet embedding GGUF files declare MEAN pooling; model card says last-token

  microsoft/BitNet#640 · Filter by author rocketlang opened on Oct 10, 2026
  1

• ؟؟؟؟؟

  microsoft/BitNet#636 · Filter by author phox9622-dotcom opened on Oct 6, 2026

• bitnet-b1.58-2B-4T: weight_scale precision (bf16 vs I2_S f32) and packed trits vs the bf16 master weights

  microsoft/BitNet#632 · Filter by author dmitrii-f-t27 opened on Sep 23, 2026
  1

• build/bin/llama-cli throws SIGSEGV, Segmentation fault when using a TL2 GGUF model

  microsoft/BitNet#631 · Filter by author hbaastrup opened on Sep 16, 2026
  3

• Verify evals on Papers with Code

  microsoft/BitNet#630 · Filter by author NielsRogge opened on Sep 14, 2026

• Latent int16 overflow in ggml_vec_dot_i2_i8_s_1x1: 128 maddubs results per lane against a 32767 ceiling

  microsoft/BitNet#629 · Filter by author purpleskulll opened on Sep 13, 2026
  5

• i2_s AVX2 GEMM folds the int16 accumulator 10x more often than its own bound requires (7.5% measured)

  microsoft/BitNet#628 · Filter by author purpleskulll opened on Sep 13, 2026
  5

• Regression: Falcon-E models cannot be loaded — "unknown pre-tokenizer type: 'falcon_e'" (support added in #268 was lost in the llama.cpp submodule update)

  microsoft/BitNet#622 · Filter by author claudioaldrighettiatroos opened on Sep 6, 2026

• convert-hf-to-gguf-bitnet.py --outtype i2_s silently writes F16 (not I2_S) for LlamaForCausalLM BitNet checkpoints (Falcon3 / Falcon-E 1.58bit)

  microsoft/BitNet#621 · Filter by author claudioaldrighettiatroos opened on Sep 6, 2026

• README "Convert from .safetensors" flow is broken: convert-ms-to-gguf-bitnet.py KeyError MODEL_ARCH.BITNET_25, and llama-quantize has no I2_S ftype (setup_env.py i2_s conversion path also affected)

  microsoft/BitNet#619 · Filter by author claudioaldrighettiatroos opened on Sep 5, 2026

• ggml-cpu.c: src1_cont undeclared → build fails on ARM CPUs with i8mm (Apple M2/M3/M4, Graviton3, ...)

  microsoft/BitNet#618 · Filter by author claudioaldrighettiatroos opened on Sep 5, 2026
  2

• [Bug]: I2_S GEMM fast path produces garbage for multi-token prompts on AVX-only CPUs

  microsoft/BitNet#617 · Filter by author puddintaim1975-dev opened on Aug 26, 2026
  1

• setup_env.py on MacBook M2 failed

  microsoft/BitNet#611 · Filter by author YHX2010-Dev opened on Aug 14, 2026
  1

• Model Hallucination Occurs again and again.

  microsoft/BitNet#609 · Filter by author dev-sohaib-asif opened on Aug 11, 2026

• Current bitnet-b1.58-2B-4T checkpoints contain all-zero MLP weight tensors (model generates garbage)

  microsoft/BitNet#608 · Filter by author grave0x opened on Aug 11, 2026
  3

• bitnet-b1.58-2B-4T runs FFN with SiLU instead of relu2 - wrong logits on every backend

  microsoft/BitNet#602 · Filter by author msrovani opened on Aug 8, 2026
  3

• i2_s SIGSEGVs at n_ubatch >= 32: BLAS backend dequantises by a row stride 4x the real packed row

  microsoft/BitNet#601 · Filter by author EUPHEMEME opened on Aug 6, 2026
  2

• BitNet-b1.58-2B-4T produces garbage output on ARM64/NEON (no AVX2) — scalar fallback uses wrong I2_S unpacking scheme

  microsoft/BitNet#600 · Filter by author gu520hz opened on Aug 5, 2026
  7

• I2_S produces incorrect output on aarch64

  microsoft/BitNet#598 · Filter by author kiljoy001 opened on Aug 3, 2026
  4

• [Windows] Linker error "undefined symbol: quantize_i2_s" when building with -q i2_s or -q tl2 on Windows 11

  microsoft/BitNet#595 · Filter by author Jimmy-sheep opened on Jul 26, 2026

• [Bug]: BitNet FFN uses SILU instead of ReLU² — perplexity 99.8 vs 17.1 on every CPU backend

  microsoft/BitNet#588 · Filter by author hmeiland opened on Jul 19, 2026
  11

• i2_s ARM CPU kernel uses wrong weight-group layout (64/16 vs 128/32) → garbage output on all ARM CPU inference

  microsoft/BitNet#585 · Filter by author EUPHEMEME opened on Jul 15, 2026
  2

• deosnt compile and run out of the box

  microsoft/BitNet#578 · Filter by author mughalh opened on Jun 28, 2026