id: deepseek-ai__DeepSeek-V4-Flash-0731__strategic
status: collected
name: DeepSeek-V4-Flash-0731
org: deepseek-ai
source:
  kind: huggingface
  url: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
  revision: 7872f01b1d1fe23eabc4c98b48bffcef5a386062
  model_card_url: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
  project_url: https://www.deepseek.com/
  paper_url: https://arxiv.org/abs/2606.19348
license:
  spdx: MIT
  url: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
  allows_offline_retention: true
  allows_local_ops: true
  allows_fine_tune: true
  notes: "Official card: repository and weights MIT. Confirm at pin."
profile:
  summary: "S-tier open MIT MoE — 284B base / ~13B active (304B with DSpark); not a 12B dense model."
  original_source: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
  use_cases:
    - "Primary open-weight agentic / coding identity after Jul 2026"
    - "Offline optionality if DeepSeek API access changes"
    - "Future multi-GPU serve (vLLM DSpark / SGLang); not current R-spine"
    - "Capability bar for daily briefs against closed frontier"
  sweet_spots:
    - "Best MIT-licensed open near-frontier that actually fits cheap Glacier (~167 GiB HF tree)"
    - "Fits the €15/month Scaleway budget with room for K3"
    - "Displaces DeepSeek-V3 as the collectable S1 identity"
  not_ideal_for:
    - "Current single-consumer-GPU daily ops (use R spine / Qwen3.8-27B)"
    - "Treating 'Flash' as a 12B dense model (that reading was an executor error)"
  capability_notes: >
    Official V4-Flash release (2026-07-31). Same structure as
    DeepSeek-V4-Flash-DSpark (speculative decoding module attached).
    HF card reports 304B params (284B base + draft). ~13B active.
    1M context. Terminal Bench 2.1 82.7. HF tree listed ~167 GB.
    Automated briefs 2026-08-04…14 repeatedly called this a 12B dense
    consumer-GPU model and once claimed it was already collected. It
    was not. Size class is S, not R.
swot:
  strengths:
    - "MIT, official org, actually downloadable"
    - "Near-frontier agentic scores at ~13B active"
    - "Footprint (~167 GiB) is cheap on Glacier and was always inside the old 850 GiB quota"
  weaknesses:
    - "Serve needs multi-GPU / high unified memory, not T1–T2"
    - "Executor hallucinated it as 12B for two weeks — catalog must stay the authority"
  opportunities:
    - "First S pull onto Scaleway (FI-WP-0004-T09)"
    - "Compressed / GGUF community packs exist if we want a later R experiment (not a substitute for the official tree)"
  threats:
    - "Successor V4.x could land while we still have no object on disk"
    - "Quota-era deferral already cost a month of optionality"
size:
  total_bytes: 166898547054
  total_human: "166.9 GB / 155.4 GiB; 67 selected files at pinned commit, HF metadata 2026-09-14"
artifacts:
  - path: "LICENSE"
    sha256: "f2c6c602815669d292889e5be8c802f2ed950653b77999b1584e8e6aed25d040"
    bytes: 1084
  - path: "README.md"
    sha256: "252acafdc9204d0dba3fde1b0a93d71cd1664a4ceadfe222b60117ed0ccc56ff"
    bytes: 7238
  - path: "config.json"
    sha256: "6c8f3d2d3b48707541b88f32f22ef3f0f8a6b57d8523281e2b8d3cdb0ae9a023"
    bytes: 1888
  - path: "encoding/README.md"
    sha256: "294ed6493ebfe5a487f172154d60c494899997fed8806fdd3352dd8e3c33f9de"
    bytes: 9168
  - path: "encoding/tests/test_input_1.json"
    sha256: "10e0c074c977c3a80daab758af28219c6b1c2bd7f3f5cf2890c84b361cc32897"
    bytes: 2748
  - path: "encoding/tests/test_input_2.json"
    sha256: "c44ae0db20fafff38a6021e8068d7ed6e28605d76cafd20be80b03398509f447"
    bytes: 526
  - path: "encoding/tests/test_input_3.json"
    sha256: "37bf8ef95e0411ea5f411be0b02fbafec7363438b6ccefddca0c52ec9aeaf69a"
    bytes: 4546
  - path: "encoding/tests/test_input_4.json"
    sha256: "c45bbd0a1b7a2f75033d8db4ba74ee5b7653bd01114045a52f79d3b387663465"
    bytes: 2730
  - path: "encoding/tests/test_output_1.txt"
    sha256: "9b366d9d2eac842a6e890594aac0b58648e5623717202b33497afadf03e26540"
    bytes: 2390
  - path: "encoding/tests/test_output_2.txt"
    sha256: "ca66b01a1ac3a204bb032c928fb607d4877171e998a50f7f89f39fa821b75665"
    bytes: 342
  - path: "encoding/tests/test_output_3.txt"
    sha256: "b3b1cd8748b7b90d3c6be6da3f786f12e4d70be073bd445ea162dfad4dc01a64"
    bytes: 3313
  - path: "encoding/tests/test_output_4.txt"
    sha256: "60e1643840ba9e4aeede450feb7b0498fa66ee24e4a939d48855ce04ec6fc375"
    bytes: 2552
  - path: "generation_config.json"
    sha256: "5fccff80f55a4d455bbe516bdd552edf3e9623df95e99fbf2a3c3389fdf91af0"
    bytes: 170
  - path: "inference/README.md"
    sha256: "b9fe8027f91f2160b46582b61eaaf7b9873b3da36ee9f98bc76a900e7c1fed94"
    bytes: 951
  - path: "inference/config.json"
    sha256: "c90861f3d10a9e4ef5954f8f1a34c529d480da1c5799f84660028f4e38e14e71"
    bytes: 1162
  - path: "inference/requirements.txt"
    sha256: "857e0b8b58e41cabe16e55bf4ab7ff791677c53b25f0f3e104ef85227cd11eab"
    bytes: 92
  - path: "model-00001-of-00048.safetensors"
    sha256: "f3668ba4cccf1ca6a7eb84e888fb92c1cdc7204d472ba9db771e6fd3abf6b874"
    bytes: 1059061856
  - path: "model-00002-of-00048.safetensors"
    sha256: "77b26c939a0e25b3113c8d6bb04e1901a748bd4a7d2589e3bfdaabdf1e9bba14"
    bytes: 3566321192
  - path: "model-00003-of-00048.safetensors"
    sha256: "412abf4c906faadc221ef0cb50f90fe20bde8454a08ad4dc2364b6b79e7fda5c"
    bytes: 3566321192
  - path: "model-00004-of-00048.safetensors"
    sha256: "9610f56bc587fb0ff9a8b68a60299482ee8c433fe5b5587e4257aca98add4a2e"
    bytes: 3596229272
  - path: "model-00005-of-00048.safetensors"
    sha256: "f87a5ac7b8becc31f9c3169afd3a6f33fb82b4af9e21022e3755a10bc28f0180"
    bytes: 3568768976
  - path: "model-00006-of-00048.safetensors"
    sha256: "4a4f3764e3fc772b9fba67f0a44ef68e18f178b6f00faa80b75db549e51894cd"
    bytes: 3590024776
  - path: "model-00007-of-00048.safetensors"
    sha256: "df81bb80e27a689e01fa579eebd6499f86e0b6105f7fea18961aa5eebbbee9bc"
    bytes: 3568768976
  - path: "model-00008-of-00048.safetensors"
    sha256: "224968d2b27f8669365ec08657a768dfec40da0585f85f302a31495931f6a526"
    bytes: 3590024776
  - path: "model-00009-of-00048.safetensors"
    sha256: "04d69ef1071fff8721c62968c200a5583122b59b015e9ef9b2978bfed271b2b7"
    bytes: 3568768976
  - path: "model-00010-of-00048.safetensors"
    sha256: "627145f4ebeb1cc3f5bdd03416b8cb7370b3c96974853cbeb8e5516ad5713e49"
    bytes: 3590024776
  - path: "model-00011-of-00048.safetensors"
    sha256: "e4b8e601dcbebe902e0102e7b098b670a121cb8b9564dd719fc41d782c8416e0"
    bytes: 3568768976
  - path: "model-00012-of-00048.safetensors"
    sha256: "64ed4e5f6126ba029c462c9d5fca0fc907c5f855b4ba01194d79560f6db16e42"
    bytes: 3590026352
  - path: "model-00013-of-00048.safetensors"
    sha256: "8dfe199d07c07ddd141c2c0136a2237f1161250a1a03ebe8deaabac93440da1d"
    bytes: 3568770544
  - path: "model-00014-of-00048.safetensors"
    sha256: "45db2f540f825f92453c50335e49aede58cca56bc578d1787c12a0fbca6593e5"
    bytes: 3590026352
  - path: "model-00015-of-00048.safetensors"
    sha256: "5810381a0f05b7381c002d299ed6ac19e42eba8070dd17e2703546944d84f292"
    bytes: 3568770544
  - path: "model-00016-of-00048.safetensors"
    sha256: "e0530b7024771b0ce2df9b40bcc2232578f3300178487ec216863b0b2835617b"
    bytes: 3590026352
  - path: "model-00017-of-00048.safetensors"
    sha256: "ed11130247118b185ade893c0109bad896dd394cb1e066ce4fce044176261d94"
    bytes: 3568770544
  - path: "model-00018-of-00048.safetensors"
    sha256: "e393fea96da2a3414ef089354fc32e1c8891954de40958c84d2c2ecf80365b25"
    bytes: 3590026352
  - path: "model-00019-of-00048.safetensors"
    sha256: "a74ca4d3e8e82ce20c458bb8b1900110b753793ad4c58d08f38995f719c616f7"
    bytes: 3568770544
  - path: "model-00020-of-00048.safetensors"
    sha256: "9f556769926e60309e8defe45ab59fc8b26ae460d30c190cd746a3d78c11e2c2"
    bytes: 3590026352
  - path: "model-00021-of-00048.safetensors"
    sha256: "1671cce7f90d781f796b5ca6bf32dd1aeb740abcb2735e41ffd28f62485ce005"
    bytes: 3568770544
  - path: "model-00022-of-00048.safetensors"
    sha256: "decd67a4bd97a75fa36861d2ad3067afeefa6a04a20da997fe6c19f171e70132"
    bytes: 3590026352
  - path: "model-00023-of-00048.safetensors"
    sha256: "c61a3e179cdbee19bfb8cbc4e111928ce2f1e1f0f4729d7c0cd5634354a4689d"
    bytes: 3568770544
  - path: "model-00024-of-00048.safetensors"
    sha256: "fc27aeb4233534f6f7781dcfe57127a3908ae10fc025c5d86dc0682057f8b2fe"
    bytes: 3590026352
  - path: "model-00025-of-00048.safetensors"
    sha256: "a66b6b8d5821b68f5b511e4f91e12025cd07d0fa6d0b71e722d825a2d6d878ca"
    bytes: 3568770544
  - path: "model-00026-of-00048.safetensors"
    sha256: "657b89314fbaf6eee4acce24b3baf7e5fd2c5986a96ad85b08d90539cde869fe"
    bytes: 3590026352
  - path: "model-00027-of-00048.safetensors"
    sha256: "fb01f21a0da0446b0bdf25a127ab19a6b06006acd8735f06a9ebfe34423fd7f5"
    bytes: 3568770544
  - path: "model-00028-of-00048.safetensors"
    sha256: "b2fd5cbbb639f16e673bc484e5cca16b52a58bf2ab4bd62592e0c5408712ad7c"
    bytes: 3590026352
  - path: "model-00029-of-00048.safetensors"
    sha256: "9ec2fdf900275daeac0980490c5c731cc7868b151ce1de5698f48418de4fa5f0"
    bytes: 3568770544
  - path: "model-00030-of-00048.safetensors"
    sha256: "9ed3c317bf967d32133ad3a068ee4c56aae9784bc8b7da694482437f37dc1782"
    bytes: 3590026352
  - path: "model-00031-of-00048.safetensors"
    sha256: "d5078c3fca3e6370043606ead7856e0b8fe67a9aab52c415769f29934c4d7f5d"
    bytes: 3568770544
  - path: "model-00032-of-00048.safetensors"
    sha256: "163653848f002718d3deaa6ce48885483fc1f2e12e50e44a47477c73ccd91393"
    bytes: 3590026352
  - path: "model-00033-of-00048.safetensors"
    sha256: "f2cffd43f2a5f491f4691f8694e6bc08239158e143ae7063dc04f0eb0259214a"
    bytes: 3568770544
  - path: "model-00034-of-00048.safetensors"
    sha256: "0f94945121474cfdb6a9ab175914d3811ffbf08e6cc54082e14e473c755d18d8"
    bytes: 3590026352
  - path: "model-00035-of-00048.safetensors"
    sha256: "9cb6a316989f7c7385e3ec2bd42ffe766ee126c70ef3466742849982ee1b0f0f"
    bytes: 3568770544
  - path: "model-00036-of-00048.safetensors"
    sha256: "7e6761421fe944c2143eb897b983085891b421c148c6c17fb5cd8eaa9bdaa497"
    bytes: 3590026352
  - path: "model-00037-of-00048.safetensors"
    sha256: "a59d662f1143596d56c452a1230b717ce43edf678207398c573a2503b0f72c91"
    bytes: 3568770544
  - path: "model-00038-of-00048.safetensors"
    sha256: "137fa617a74ba8e73fd76bb1010c7a85d791aaa150006cb66faa04a83e9e730f"
    bytes: 3590026352
  - path: "model-00039-of-00048.safetensors"
    sha256: "a29af1aa519d7ce726235ea2c2b38146d756290cda7a82d90c4d4438155b53e4"
    bytes: 3568770544
  - path: "model-00040-of-00048.safetensors"
    sha256: "8bc93d8a7d1987dc86b14e22b1d8f42ec31da92c56edd9f312daf43f33a6a206"
    bytes: 3590026352
  - path: "model-00041-of-00048.safetensors"
    sha256: "fd312e7fdd6cb5796df356a7f0314f124851dc149991e9fb02c5bed45cc4ba05"
    bytes: 3568770544
  - path: "model-00042-of-00048.safetensors"
    sha256: "4d19bf368083c9a183cb0849f316ec17b62f859ca824c0586e779657efb6e6a6"
    bytes: 3590026352
  - path: "model-00043-of-00048.safetensors"
    sha256: "b7103842ceb70848f9804f55c193d6a57f43174a587cba42b61b5c1bc4e1303d"
    bytes: 3568770544
  - path: "model-00044-of-00048.safetensors"
    sha256: "422d3889fa20c238b7f97464c14df0bcf3328f189c294f41a3a334421dc560c7"
    bytes: 3590026352
  - path: "model-00045-of-00048.safetensors"
    sha256: "a5be6aed7b84fc87ec42b5d24ba0b0d67f253a3906fcd99c13f4f7be5958fc00"
    bytes: 1059332516
  - path: "model-00046-of-00048.safetensors"
    sha256: "5db924ca907e0d93acd975bd5079c3662717f9ac709f23d079bd8f816d29d9dd"
    bytes: 3610455184
  - path: "model-00047-of-00048.safetensors"
    sha256: "62816173f9f6e136b20b48e3b6f16613ac9ea02b5603f636928b253244a548bd"
    bytes: 3560111960
  - path: "model-00048-of-00048.safetensors"
    sha256: "cc43742bd24ae6bcdea343a91442f6f66aed2cfebcc6b235470204851ce2f8a9"
    bytes: 3692775244
  - path: "model.safetensors.index.json"
    sha256: "98efab455cf08dfbbbaaba6f570e1bf10bf927d2b4c3c453a59c2f6f0e3be92b"
    bytes: 5602871
  - path: "tokenizer.json"
    sha256: "8f9f37ca37fdc4f5fd36d5cf4d3b0e8392edb4e894fd10cc0d70b4957c8633cf"
    bytes: 6367146
  - path: "tokenizer_config.json"
    sha256: "6ac8c8dc065ed118161d02dd532749ae3f52c243deac27872134fae2f50d8547"
    bytes: 801
hardware_class:
  min_vram_gb_q4: 0
  min_vram_gb_fp16: 0
  notes: "Beyond current lab run envelope (T3–T4). Community 2-bit reports exist for 128 GB unified memory; not an R-spine claim."
axes: [A, B, D]
priority: high
collection:
  approved_by: "grok"
  approved_at: "2026-09-13"
  downloaded_at: "2026-09-13T23:18:26Z"
  downloaded_by: "fi-reserve-collect (railiance01)"
  storage_path: "s3://railiance-fi-open-weight-reserve/strategic/deepseek-ai__DeepSeek-V4-Flash-0731/7872f01b1d1fe23eabc4c98b48bffcef5a386062/"
  manifest: "s3://railiance-fi-open-weight-reserve/manifests/deepseek-ai__DeepSeek-V4-Flash-0731/7872f01b1d1fe23eabc4c98b48bffcef5a386062/MANIFEST.json"
  brief_refs:
    - briefs/2026/08/2026-08-03.md
    - briefs/2026/09/2026-09-13.md
    - docs/decisions/2026-09-13-scaleway-object-reserve.md
reason: "S1 collectable — strongest MIT open MoE that fits the Scaleway euro budget; first pull after bucket create."
tags: [tier-s, strategic, moe, mit, beyond-run-envelope]
companions: []
notes: >
  Do not confuse with a 12B dense checkpoint. There is no
  deepseek-ai/DeepSeek-V4-Flash-0731-12B. Pull the official card repo.
  Storage class at upload: strategic/ → Glacier.
history:
  - at: "2026-08-03"
    event: nominated
    by: grok
    detail: "Catch-up brief named V4-Flash-0731 as mid-tier open value leader."
  - at: "2026-08-04"
    event: misidentified
    by: rein-aharness
    detail: "Automated briefs treated it as 12B dense; never cataloged; 08-07 claimed on-disk collection that did not happen."
  - at: "2026-09-13"
    event: approved
    by: grok
    detail: "Corrected to 304B-class MoE ~167 GiB MIT. Approved for Scaleway Glacier once FI-WP-0004-T08 bucket exists."
  - at: "2026-09-13"
    event: collecting
    by: grok
    detail: "T08 bucket live. snapshot_download + Glacier upload started on workstation staging /home/worsch/vault/freedom-intelligence."
  - at: "2026-09-14"
    event: collecting
    by: codex
    detail: "Replaced WSL snapshot staging with diskless Railiance service fi-reserve-collect. 64 MiB parts, 512 MiB memory cap, no swap. Source pinned to 7872f01b1d1fe23eabc4c98b48bffcef5a386062; 67 selected files / 166898547054 bytes. Destination strategic/deepseek-ai__DeepSeek-V4-Flash-0731/<commit>/. Complete manifest pending; not collected or restore-verified."
  - at: "2026-09-22"
    event: collected
    by: claude
    detail: "Transfer finished 2026-09-13T23:18:26Z; complete MANIFEST.json published. Re-checked 2026-09-22: 67/67 objects, 166898547054 bytes, all GLACIER; 48 LFS SHA256 source digests equal computed SHA256, 19 git-sha1 files verified in-stream; HEAD size/ETag/version match manifest for every object. No restore readback yet, so not verified."