Sujets / Setup & install
Setup & install
Auto topic: Setup & install
Issues
- #6281 [Engine] SSD conversation tier turns off when MTP comes from a GGUF (fingerprint looks for mtp/rt/dense.bin)
- #6542 [Engine] Findings on Apple M5 Max 128 GB: Gyro-L and Hy4-Flash for coding agents on v0.2.18 (+ PRs #6538, #6539, #6540)
- #5949 [Engine] HIP/Linux: make HSA_USERPTR_FOR_PAGED_MEM=0 the default on discrete AMD cards
- #5939 [Engine] Windows CUDA12 0.1.59: Tesla V100 GPU is lost during single- and dual-GPU inference
- #6531 [Benchmark] gfx1100 (RX 7900 XTX, IQ3_S): existing switches give 1.9–3.7x prompt and 1.2x decode speedup
- #6521 [Benchmark] R9700 + Threadripper PRO 3975WX: Hy4-Flash at ~30 tok/s (SQ-64, 4K) vs. 62 tok/s in the 0.2.18 notes
- #6523 [Engine] Windows NUMA in Strata: Why Our 40-Core Server Initially Used Only 20 CPU Workers
- #6526 [Engine] gfx1200 (2x RX 9060 XT, ROCm 10.1): peer-GPU fused WMMA prefill, RDNA4 MMVQ load_int_b2/Q6KTraits, and hipBLASLt fallback notes
- #6438 [Engine] Strata 2.17 + Lazarus 4.8 + Lazarus MCP Assistent not working ?
- #6525 [Engine] Drop in MTP head replacement trained against quantized model output: +3.7 / +4.4 % decode at --spec 3
- #6339 [Engine] 0.1.62: --lookup-max-t past 8 with --draft dflash2 ends the request (dflash2: draft rows out of range); a small fix, +11..13 % on file rewrites
- #6518 [GUI] Model picker shows "Download size" for a fully downloaded Qwen-family model (IQ3_S); Gyro-S in the same folder shows "Already on this computer"
- #6435 [Benchmark] RX 9070 (gfx1201, Windows 11, 32 GB RAM) / Gyro-S TQ1_0 / 0.2.6 → 0.2.17: 0.2.17 defaults lose ~30% prompt and ~12% decode on a 32 GB PC; two settings bring it back
- #6516 [GUI] Subagent list: log dialog closes the popover; helper work and replies are not visible
- #6517 [Other] Feature request: rename-aware editing, an integrated debugger, and an optional reviewer model
- #5629 [Engine] GPU fatal error: "layer X never rang (an illegal memory access was encountered)"
- #6489 [Engine] 16K cold prefill regression on a single RTX 4090 D (sm_89, IQ3_S): 0.1.55 → 0.1.60/0.1.63 (−21 %), unrelated to --prefill, --expert-cache or the route tail skip
- #6273 [Engine] 16 GB cards: guard the ~1 GB VRAM floor during KV growth and document the per-format context limit
- #6232 [Engine] 384K compaction re-reads 264K tokens (456 s): content-addressed, position-independent segment cache for near-zero-reread compaction
- #6509 [Engine] Engine 0.1.55+ decodes 35–55% slower than 0.1.52 on RTX 3080 Ti + RTX 4070 (Windows): decode expert-cache hit falls from ~92% to 55–70% although the cache gets ~40% more slots
- #6507 [Engine] q36: one early byte change costs the whole prompt — all-or-nothing match_len has no GDN state to step back to (0 reused + 38026 read, 108 s)
- #5839 [Engine] 0.2.13 quoted EOS still ends agent turns with reasoning-only stop, with opt-in recovery results
- #6500 [Engine] Vulkan backend: scope for the port (stage 1 landed: the loop and one Q4_0 kernel)
- #6370 [Engine] Engine does not update to latest one
- #5941 [TUI] Terminal does not have "speed projection/uncensoring" on/off toggle like earlier .bat based configuration had
- #6327 [Engine] Compaction re-reads 331K-342K tokens five times (19.5-20.1 min each): the cache budget is the blocker (PR #4338)
- #6496 [Other] Windows HIP rdna2 package (0.1.63) works on a real RX 6900 XT (gfx1030)
- #6491 [Other] Ray, an assistant and coding agent built on Strata: would any part of it help the app?
- #5787 [Engine] serve: python.exe dies natively under long agent sessions — every dump faults inside CPython on the same nested `any(genexpr)` path (`serve/frontend.py` `_has_tag`)
- #6409 [Engine] Qwen3.8-Next-Flash on Strix Halo AI Max 395 (128GB UMA): tuning for 2 concurrent streams at >100 tok/s each @ 200k context
- #6472 [Benchmark] RTX 4090 24 GB VRAM headroom for long-context KV and IQ4 experts
- #6404 [Engine] gfx1030 (RX 6800 XT, Windows): 0.1.63 reads prompts 2x faster than 0.1.60 but decodes 11 % slower — same config, A/B on one host
- #6470 [Engine] CUDA Q8_0 mmproj loses its VRAM saving with the native picture reader
- #6469 [GUI] The desktop agent's compaction request re-reads the conversation from the system prompt on Gyro (hybrid); --prompt-cache-spread cuts it from 14.9 to 2.7 s here
- #5672 [Engine] gfx1100 (RX 7900 XTX) --adapt-async 1 results in timeout followed by a display-driver reset during long generations
- #6429 [Benchmark] hipBLASLt tuning table by default on gfx1100 (RX 7900 XTX, IQ3_S): speed, decode and logits
- #6462 [Benchmark] RTX 4090 (sm_89): sm_86 bit-exact set +6%, sm_120 switches +4% on UD-Q4_K_XL, CPU share 32/6144 cuts 3K-token follow-up turns 26%
- #6460 [Engine] Claude Code auto-compaction returns an empty summary near the context limit unless `anthropic_thinking` is `on_request`
- #6448 [Engine] Scope: plan an integrated Intel GPU (Arc B390, Panther Lake) as one pool of memory with the system in the SYCL engine
- #5678 [Engine] Qwen3.8-Flash-Next IQ3_S on sm_121 (unified memory): prefill stalls at 33,372 of 33,379 tokens, just past the 32,768 chunk; STOP not honoured, restart never finishes
- #5917 [TUI] add custom access IP option
- #5990 [Engine] V100 (sm_70) on Linux: 13× Xid 79 + PCIe CmpltTO across engines 0.1.35–0.1.59.1 — not version-specific
- #6208 [Engine] 0.2.15 decodes 28-36% slower than 0.2.7 on RTX 3060 12 GB: --expert-cache auto starts with a much worse resident set (same pack and profile)
- #6444 [Other] Linux - Build failed after update - .venv move to terminal broke pip
- #5995 [Engine] Exit code 0xC0000409: stack buffer overrun on V100 + RTX A4500
- #6118 [Engine] Cold 50K prefill regression: 0.1.52 → 0.1.59 → 0.1.60 on RTX 3060 12GB (−16.5%)
- #6426 [Engine] --layer-split auto with DFlash2 on two cards: the noisy PCIe ratio around 4.0 picks a placement with no expert room, and the drafter is dropped
- #6423 [Engine] 0.2.17 sm_120 defaults regress prefill and cache-reuse TTFT on a mixed-arch layer split with the 12.0 card first
- #6252 [Other] Intel Arc (xe driver): the installer fallback always uses --stream-experts, even with enough host RAM for a resident arena
- #6417 [Engine] 0.1.63: after one picture through the elastic image encoder, the expert cache can grow back a segment short and --prefill auto:32768 sticks at 16384 (60K cold prompt -37 %)
- #6175 [Engine] the 4-bit KV cache (q4u32) as the default on RDNA4 — scope agreement
- #6413 [Benchmark] v0.2.17 on 2x Arc Pro B60 (layer split) and 1x Arc Pro B70 / Qwen3.8-Flash-Next UD-Q4_K_XL / 262K: same speed as v0.2.16, `--pipeline-windows 2` no longer produces `!!!!` (12 of 12)
- #6392 [Engine] Linux HIP RX 7900 XTX (32 GB RAM, IQ2_XS low-RAM): engine hangs within 4-10 min of sustained serving, verify timeouts, also on 0.1.63
- #6244 [GUI] Changing an option on the model screen silently drops drafting: dflash2 and resets STRATA_RESIDENT_PIN to 0
- #6405 [Engine] 0.2.17: on a 3-card layer split the expert cache shrinks from 5,940 to 3,551 slots (10.40 → 6.19 GiB) at the same VRAM footprint
- #6400 [Engine] + [Docs] SYCL setup & LXC/multi-GPU flow is a complete mess (paths, build dir, device selection)
- #6250 [Engine] Arc Pro B70 (xe): server reaches "ready", first request dies with verify flag A wait
- #6397 [Engine] Engine cannot use more than 8 videocards
- #6402 [Engine] serve: /v1/models reports the context window but no output-token limit - clients guess the answer cap
- #5756 [Engine] Tool-call value truncated when the model skips a parameter's own </parameter> after a nested pair (Qwen3.8-Flash-Next)
- #6269 [Engine] gfx103x: prompt throughput gaps left by the MMQ removal (dense projections, IQ4_XS down)
- #6342 [GUI] agree scope for desktop agent continuity and context controls
- #6254 [Engine] Cold-boot first load of Qwen3.8-Flash-Next-Gyro-M crashes with 0xC0000005 on RX 9060 XT (gfx1200)
- #6203 [Engine] serve/server.py engine_args() ignores hand edits to args without user_args
- #6174 [Docs] Mac: source setup gives Q2_0 the DFlash2 drafter, but MAC.md says only the two Unsloth sets get it
- #6173 [Other] Source setup accepts --context 262144 for Qwen3.6-35B-A3B on a 12 GB card, and the engine then refuses to start
- #6172 [Other] Source build fails when sccache is on PATH: ggml's global RULE_LAUNCH_COMPILE wraps Strata's nvcc compiles
- #6171 [Engine] 0.2.15 window warm-up with STRATA_ROUTE_PRIOR: the first request after a start often ends in the middle of a number
- #6130 [Engine] RX 7900 XTX (gfx1100, Linux), 0.2.15: picture text misread ("STRATA 412" for "STRATA 42") after earlier requests on the default HIP build; the -DSTRATA_MMQ=ON build reads it right
- #6128 [Engine] 0.1.60 on Windows: with --vram-reserve-mib 700 the expert cache grows past the auto size and decode drops to 0.2-42 tok/s on an RTX 4090
- #6111 [Engine] STRATA_OWN_CPU_QUANT=ON (0.2.15 x86_64 default) is ~4% slower than ggml-cpu on AVX2 CPU expert pool decode (IQ3_XXS)
- #6110 [Engine] explicit --layer-split reloads CUDA0 dense weights a second time at startup (0 MiB freed)
- #6086 [Other] Terminal: shell tab completion for the strata command
- #6019 [Engine] Rare non-finite logits (#879) with native IQ2_XS and IQ3_XXS packs on sm_86, engine 0.1.59.1 (none with Q2_0 on the same flags)
- #6325 [Engine] RDNA4 16 GB + 6-core CPU, Gyro-S: the automatic PCIe share (0.90) is ~3 % slower than --pcie-frac 1.0; with --spec 6 --draft-conf 0.6 decode is +7.5 %
Pull requests
- #4310 [Engine] session files: format v2 with per-stage save/restore under --layer-split
- #6520 [Engine] Inkling-2-Mini from the official NVFP4/MXFP8 checkpoint, expert offload and CPU experts for 8-16 GB cards
- #6315 [Engine] serve: a tool call written in a shape its schema does not declare is read as the call it meant (#6285)
- #6210 [Engine] Opt-in P2P q8_1 decode push, fused P2P down-projection, and GPU peer sync flag
- #6548 [Engine] Add opt-in streamed-KV QSA slot-group batching, pre75 multi-slot attention, and batch SIDE_MIXER
- #6451 [TUI] strata launch --model-path: name Swift's shard 1 as --ple-gguf, where its n-gram table is
- #4878 [GUI] Keep quiet desktop chat streams alive
- #6305 [Engine] fingerprint the resolved MTP files, so the SSD conversation tier stays on for GGUF and rt drafters
- #6540 [Engine] Hy4-Flash: SQ-96 starts on 128 GB Macs, --info no longer crashes, session reuse on by default, reasoning_effort max in configs
- #4411 [Engine] session-file saves: ping-pong blocks and a writer thread so the source copy overlaps the previous block write
- #4307 [Other] serve+webui: unload that works — pre-staged checkpoints, polite waits, honest answers
- #4306 [Other] serve: session ladder — conversations survive an engine unload
- #3553 [Engine] let a HIP helper read pinned expert rows in place
- #3552 [Engine] skip file-tier experts while a helper GPU is attached
- #3551 [Engine] add an opt-in CUDA and HIP hybrid build
- #4551 [Engine] session-file v3: incremental append saves under --session-save-append
- #6532 [Engine] prefill: opt-in STRATA_I8_DEFER - unpicked streamed entries no longer cut the int8 expert groups into fragments (gfx1030 4K prompt +3.6%)
- #3800 [Engine] Token mask for JSON response formats: MTP drafts survive constrained decoding
- #5436 [Engine] opt-in STRATA_ARENA_NUMA: place the expert arena per NUMA node on two-socket Linux
- #6506 [Engine] add Inkling 2 model family and dual-GPU looped MoE engine support
- #3607 [Engine] native Windows support for Intel Arc SYCL, rebased onto 0.2.18 (engine 0.1.64)
- #5820 [Engine] Multiple API keys with per-key concurrency limits and reserved capacity
- #6365 [Engine] Prefer FP32 TLA sparse prefill on Battlemage in builds with local TLA headers
- #6510 [Engine] --experts-wex4: read the routed experts from a checked WEX4 container (opt-in, file tier)
- #6102 [Engine] use runtime expert count in layer-major prefill
- #4376 [Engine] dsv41: the WMMA expert kernels compile on pre-Volta cards
- #6197 [Engine] int8 expert kernels: IQ4_XS as a down format
- #6176 [Engine] q4u32: the 4-bit KV cache with a fitted step as the default on RDNA4
- #4255 [Other] serve: refuse a port another program holds on 127.0.0.1, and say which program
- #6492 [Other] tests: the first VisionTempFiles class was shadowed and never ran
- #6485 [Engine] Add experimental gfx90a HIP build
- #2965 setup: a separate opt-in layers 8-33 projection vector for the original Qwen3.8-Flash-Next (any quantization)
- #5476 [Engine] gfx103x decode: opt-in one-warp MMVQ for K <= 4096 and a per-format expert layout table (STRATA_MMVQ_NW1, STRATA_EXP_TABLE)
- #5475 [Engine] opt-in int8 copies for the persistent hc read (STRATA_HC_PERSIST_Q8, STRATA_HC_I8ONLY)
- #5179 [GUI] add true same-turn coding-agent steering with a reproducible Harness patch
- #1614 [Engine] shared KV pool for all lanes (--kv-pool-tokens): at 2 x 256k the first token comes 2-5x sooner and lane moves ~10x faster; off = identical to main
- #6453 [Engine] SYCL: an integrated Intel GPU is planned as one pool of memory with the system (Arc B390)
- #6452 [Engine] serve: an on-demand CPU picture reader starts without asking the expert cache for VRAM
- #6383 [Docs] add reusable project workflow skills
- #6385 [GUI] preserve completed summaries on context retry
- #6390 [GUI] restore the local agent reply ceiling at large context sizes
- #6391 [Other] verify the context patch on reused desktop Harness stages
- #6394 [GUI] preserve long-running desktop agent sessions
- #6458 [Engine] STRATA_FILL_STAGED: copy mapped blobs to the GPU via pinned buffers
- #6455 [Engine] SYCL: an Xe3 GPU (Panther Lake, Arc B390) runs the Battlemage kernels; STRATA_INTEL_GEN sets the class
- #6454 [Engine] serve: an integrated Intel GPU's memory is the system RAM in the Monitor and the free-memory checks
- #3919 [Engine] kv: size the resident ring from the run's depth (opt-in, STRATA_KV_RING_ADAPT)
- #6434 [Engine] Use eight-row tails in M1 resident expert GEMMs
- #6441 [Engine] Reduce resident Metal QSA prefill scratch
- #2427 [Engine] HC_Q8: skip the layers' bf16 hyper-connection copies on a layer split too
- #6427 [Engine] layer split auto with DFlash2: plan DFlash2's drafter and after-cache bind on the last stage, and keep the gate-off fallback loadable
- #4019 [Engine] Mac: calibrate the decode kernels' tiles on the device (setup.sh --calibrate)
- #2784 [Engine] HIP: the RX 6600 / 6650 XT (gfx1032) in setup's cards
- #6424 [Other] tests: isolate the two SYCL-path drafting tests from a real sibling Strata-data
- #4883 [Other] Setup: --gpus on Windows AMD: RDNA 2, 3 and 4 cards sharing one model
- #5494 [Other] Windows: recognise the Radeon Pro V620 (PCI id 0x73A1) as gfx1030
- #6166 [Engine] opt-in blob verifier for the direct expert path: STRATA_BLOB_VERIFY
- #6262 [Engine] IQ4_NL down rows in 512-bit lanes, opt-in, same bits as today's paths
- #6260 [Docs] Explain DFlash2 selection during Metal setup
- #6401 [Engine] oomd_pacer: discover split --native GGUF shards and keep PLE cached without --ple-gguf
- #6087 [Other] Terminal: strata complete prints bash/zsh/fish tab completion from the CLI's own grammar
- #5918 [Engine] Metal: honour --kv fp16 with direct-half KV kernels
- #5835 [Engine] Conversation cache: an opt-in disk tier for evicted, idle and oversized conversations (--conversation-cache-dir)
- #4867 [Engine] HIP/Linux: keep the kernel's NUMA balancing off the engine's memory (GPU queue evictions)
- #5346 [Other] serve: preserve multi-turn reasoning by default and skip empty historical think blocks
- #5217 [Other] Control: recommend_context counts the resident plan's expert arbitration
- #4967 [Engine] serve: warn when systemd-oomd is running, and name it when the engine is ended by signal 9
- #4865 [GUI] keep tool Approve cards under the latest reply
- #4827 [Engine] use a downloaded GGUF of another length when it is a whole GGUF file
- #4788 [GUI] Add a context gauge and conversation summaries to browser chat
- #4651 [GUI] add --resume (STRATA_RESUME=1) to the Linux AppImage so your last model loads as Strata starts
- #4582 [GUI] add --lan (STRATA_LAN=1) to the Linux AppImage so the model is on your network as Strata starts
- #4571 [Other] update.sh / UPDATE.bat: a failed pull without origin/main is not the old history
- #4295 [Engine] gfx1151 hipBLASLt table for a system ROCm 7.2.4 (hipBLASLt 100202)
- #4201 [Other] setup: sycl/setup_intel.py started with the system Python runs again with .venv's (Permission denied on /usr/.strata-pip.json)