hf2q-0.1.12

docs.rs

Rust

hf2q 0.1.12

Pure Rust CLI for converting HuggingFace models to hardware-optimized formats and serving them over an OpenAI-compatible API on Apple Silicon

Crate
Source
Builds
Feature flags

..
adr030
coherence-harness
fixtures
ggml_quants_harness
opencode-web-stack
parity
reference
testdata
adr-011-phase2-gate.sh
adr005_mteb_sanity.sh
adr010_iter64_batched_coherence_gate.sh
adr010_iter64_iter68_full_lock.sh
adr017_phase_d.sh
adr019-phase0a1-capture-and-bin.sh
adr019-phase0a1-llamacpp-capture.sh
adr019-phase0a4-perf-counters.sh
adr019-phase2-determinism-5x.sh
adr019-phase2-parity.sh
adr020_ac7_boundary_validate.sh
adr020_ac8_validate.sh
adr020_fetch_wikitext_calibration_jsonl.sh
adr022_p18_byte_equal.sh
adr027-cross-axis-sweep.sh
adr027-full-validation.sh
adr027-long-context-sweep.sh
adr028_coherence_gate.sh
adr028_full_stack_bench.sh
adr028_iter485_h4_bench.sh
adr029_aot_compare_air.sh
adr029_xctrace_per_cb_gpu_time.sh
adr034_mtp_paired_bench.sh
adr034_workload_bench.sh
aggregate-q4_0-mst.py
aggregate_decode.py
aggregate_decode_mst.py
aggregate_hypotheses.json
aggregate_hypotheses_gemma.json
bench-baseline.sh
bench-iter83-chunk-engaged.sh
bench-iter83-default.sh
bench-iter83-parse-results.sh
bench-iter8d-residency.sh
bench-long-context.sh
bench-matrix.sh
bench-needle-haystack.sh
bench-w5b10-flash-attn-prefill.sh
bench-w5b11-post-attn.sh
bench-w5b11-sunset-parity.sh
bench-w5b14-dense-q-fused.sh
bench-w5b15-dense-q-arena-reset.sh
bench-w5b17-dn-wrapper-audit.sh
bench-w5b18-qkv-split.sh
bench-w5b19-repeat-tiled.sh
bench-w5b20-repeat-tiled-mega.sh
bench-w5b21-reaudit.sh
bench-w5b22-residual-audit.sh
bench-w5b24-ffn-pooled.sh
bench-w5b26-ffn-output-lift.sh
bench-w5b27-phase-a-cold.sh
bench-w5b27-phase-a-retro.sh
bench-w5b27-phase-b-lift.sh
bench-w5b7-iter2-residency.sh
bench_embedding.sh
bench_lcp_long_prompt.sh
bench_lcp_resume_speedup.sh
bench_qwen36_agentic_cache.sh
benchmark.sh
benchmark_deepseek4_server_parity.sh
bug-gemma-cli-diverge-2026-05-16.sh
bug-gemma-extract-and-diff-2026-05-16.sh
bug-gemma-find-divergence-2026-05-16.sh
build_deepseek4_decode_cancel_request.jq
byte_cmp_full_pipeline.sh
byte_cmp_gguf.py
byte_cmp_streaming.py
coherence_and_speed_regression.sh
crawl_verify.sh
deepseek4_agentic_prompt_contract.jq
deepseek4_agentic_prompt_provenance.jq
deepseek4_agentic_request.jq
deepseek4_full_context_receipt.jq
dump_layer_states.cpp
ensure_github_release_tag.sh
exp3-tq-vs-dense-end-to-end.sh
exp4-hf2q-vs-llama-content-blind.sh
fetch_wikitext2.sh
gemma4_eight_slot_receipt.jq
hf2q_process_guard.sh
install.sh.in
install_opencode_web_stack.sh
iter37-parity-matrix.sh
iter44-bench-decode-attribution.sh
iter44-decode-bucket-aggregator.py
iter45-aggregate-results.py
iter45-chain-n-curve-bench.sh
iter61a3-bisect.sh
iter91-cb-count-probe.sh
iter91-pp4096-wall.sh
lcp_resume_mini_soak.sh
long_decode.sh
macos_memory_guard.sh
macos_thermal_guard.sh
macos_thermal_probe.swift
p11_re_emit_dwq.sh
p17b_q4k_bench.sh
parity_check.sh
peer_parity_run.sh
peer_ref_ratchet.sh
phase0_candle_bench.sh
profile-decode-mst.sh
profile-iter9-mst.sh
profile-p3aprime.sh
q8_pad_analysis.py
qwen35_bench.sh
qwen35_prefill_bucket_profile.py
qwen35_prefill_coherence_sweep.py
qwen35_prefill_warm_sweep.py
qwen35_raw_passthrough.jinja
qwen35_tokenizer_parity.sh
qwen36_watchdog_validate.sh
qwen38_long_decode_ab.sh
qwen38_matched_reference_abba.sh
qwen38_matched_reference_contract.sh
qwen38_speculation_ab.sh
reconvert-broken-2026-05-05.sh
release-check.sh
render_standalone_installer.sh
render_standalone_release_record.sh
run_agentic_cache_release_gate.sh
run_deepseek4_decode_cohort_gate.sh
run_deepseek4_matched_peer.sh
run_release_gate_process_group.sh
seal_release_binary.sh
serve_deepseek4_opencode.sh
serve_gemma4_opencode.sh
serve_qwen36_opencode.sh
serve_qwen38_opencode.sh
sign_notarize_standalone_release.sh
smoke_api.sh
smoke_lifecycle_drain.sh
smoke_openai_sdk_embeddings.sh
sourdough_gate.sh
sourdough_qwen35.sh
sunset-w5b14-legacy-dense-q.sh
sunset-w5b18-qkv-split-legacy.sh
sunset-w5b20-gqa-expand-legacy.sh
sunset-w5b24-ffn-pooled-legacy.sh
test_agentic_cache_lifecycle.sh
test_deepseek4_agentic.sh
test_deepseek4_agentic_fixture_contract.sh
test_deepseek4_cached_suffix.sh
test_deepseek4_cached_suffix_contract.sh
test_deepseek4_cooperative_prefill_receipt_contract.sh
test_deepseek4_decode_cohort_receipt_contract.sh
test_deepseek4_interactive_overlap.sh
test_deepseek4_long_context_cache.sh
test_deepseek4_opencode.sh
test_deepseek4_peer_cold_wave.sh
test_deepseek4_peer_cold_wave_contract.sh
test_deepseek4_reasoning_recovery.sh
test_deepseek4_structured_tools.sh
test_full_context_agent_slots.sh
test_gemma4_agentic.sh
test_gemma4_eight_slot_receipt_contract.sh
test_gemma4_long_short_overlap.sh
test_gemma4_long_short_overlap_contract.sh
test_gemma4_wave_thermal_contract.sh
test_getting_started_guide.sh
test_macos_thermal_guard_contract.sh
test_opencode_web_stack_lifecycle.sh
test_qwen36_agentic.sh
test_qwen36_cumulative_release.sh
test_qwen36_prefill_cancellation.sh
test_qwen36_prefill_watchdog.sh
test_qwen36_watchdog_harness_contract.sh
test_qwen38_artifact_binding_contract.sh
test_qwen38_first_image_cache.sh
test_qwen38_long_decode_receipt_contract.sh
test_qwen38_matched_reference_contract.sh
test_release_binary_seal_contract.sh
test_release_dependency_provenance_contract.sh
test_release_gate_process_group_contract.sh
test_release_tag_recovery_contract.sh
test_shipping_contract.sh
test_standalone_installer.sh
test_standalone_release_signing_contract.sh
test_web_search_fetch_plugin.sh
tq-c0-diff.py
tq-c0b-dequant.py
validate_deepseek4_decode_cancel_request.jq
verify_deepseek4_cooperative_prefill_receipt.sh
verify_deepseek4_decode_cohort_receipt.sh
verify_gemma4_parity_receipt.sh
verify_gemma4_wave_thermal_receipt.sh
verify_macos_thermal_receipt.sh
verify_qwen38_long_decode_receipt.sh
verify_release_dependency_provenance.sh
verify_standalone_candidate.sh
w49_iter120_ab_diagnostic.sh
w56_iter125_smoke.sh
wedge4_qwen3vl.sh