DeepSWE
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
Not listed as a Continuum hosted id. HF license field is other; license_name is minimax-community (see the repo LICENSE).
HF config.json (fetched 19 Aug 2026): MiniMaxM3SparseForConditionalGeneration, text stack 60 layers, hidden 6144, 128 local experts, 4 experts/token, 1 shared expert, 1,048,576 positions. Vision config present.
Use M3 when you want MiniMax’s open checkpoint. AA 45 and vals 42.72% are cited; we do not invent a DeepSWE percent.
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
Ranked view: Best coding models: the independent leaderboard puts this row and every other card that carries an independent coding score on one board, with the confidence intervals left visible.
Copied from AA or vals HTML we opened. 0.0% placeholder bars are omitted. Do not average these into the hero tiles, and do not invent CI, $/task, or steps for Artificial Analysis.
| Bench | Printed | Note | Source | URL | As of |
|---|---|---|---|---|---|
| OpenRouter list pair | $0.30 / $1.20 per 1M | Catalog row. No DeepSWE $/task: that board has no MiniMax M3 row. | OpenRouter catalog | source | 2026-08-19 |
Plus is $25/mo with $25 weekly hosted usage. The Mac app stays free with your own keys. Get Plus is not Download for Mac.
Community quants: Community quants are community.
This is the model repo, not Download for Mac. Downloads and likes from the Hugging Face API on 2026-08-19.
usedStorage and the sum of .safetensors sizes can differ (LFS pointers, non-shard files, duplicate copies). Cited separately. Source: recursive tree API, 2026-08-19.
| Path | Bytes |
|---|---|
model-00001-of-00059.safetensors | 5,583,706,344 |
model-00002-of-00059.safetensors | 10,922,058,704 |
model-00003-of-00059.safetensors | 16,079,479,008 |
model-00004-of-00059.safetensors | 16,079,479,000 |
model-00005-of-00059.safetensors | 16,079,479,000 |
model-00006-of-00059.safetensors | 16,079,479,000 |
model-00007-of-00059.safetensors | 16,079,479,000 |
model-00008-of-00059.safetensors | 16,079,479,000 |
model-00009-of-00059.safetensors | 16,079,479,000 |
model-00010-of-00059.safetensors | 16,079,479,408 |
model-00011-of-00059.safetensors | 16,079,479,408 |
model-00012-of-00059.safetensors | 16,079,479,408 |
Showing the first 12 of 59 safetensor shards. The tree also holds 23 non-shard files, none of them listed here. The snapshot was capped at 12 shards when the tree API was read on 2026-08-19, so the remaining 47 are on the Files tab, not on this page. Totals above are the full tree API sum, not this slice. Byte sizes are Hub tree API size fields, never invented.
Not listed as a Continuum hosted id. OpenRouter minimax/minimax-m3.
We list first-party and OpenRouter, plus Continuum only when the live public allowlist named the id. This is not a 15-host routing table.
OpenRouter catalog 19 Aug 2026: $0.30 / $1.20 per million. No DeepSWE $/task.
Artificial Analysis Intelligence Index 45, fetched 19 Aug 2026. The AA tile is that published integer. AA did not publish CI, steps, or $/task for this identity in a form we will copy onto the chip.
| Claim | Source | As of |
|---|---|---|
| 1.67T week tokens (+2% WoW) | OpenRouter rankings This Week | 2026-08-19 |
| OR list $0.30 / $1.20 | OpenRouter catalog | 2026-08-19 |
| HF safetensors.total 427,040,140,160; usedStorage 1,708,357,470,332; 59 shards; likes 1,480; created 2026-06-02T07:49:31Z; arXiv:2606.13392 | Hugging Face API MiniMaxAI/MiniMax-M3 | 2026-08-19 |
| OpenRouter id minimax/minimax-m3, context 1,048,576 | OpenRouter /api/v1/models | 2026-08-19 |
| HF downloads 177,676 | Hugging Face API MiniMaxAI/MiniMax-M3 | 2026-08-19 |
| 59 safetensor shards; shard bytes 854,176,398,808; tree files 82 | Hugging Face tree API MiniMaxAI/MiniMax-M3 | 2026-08-19 |
| Spaces API returned 45; official collection MiniMax-M3 (4 items, 22 upvotes) | Hugging Face Spaces / collections API | 2026-08-19 |
No other card in this cluster shares a published DeepSWE official row we can put next to this one. We will not compare on vendor-blog numbers.
Continue through MiniMax's model family with MiniMax M2.7.
The official DeepSWE board updated 13 Aug 2026 did not list MiniMax M3. Omitting is the rule.
Lab posts pick a harness, an effort, a split, and sometimes a private eval set. DeepSWE publishes the official mini-swe-agent row with cost, tokens, and steps. Scale SWE-bench Pro only counts when the public shared-harness board has a row we can fetch. A higher lab number is usually a different test, not a better one.
No Continuum hosted id is verified for this model, so this page does not invent a curl target. Use the first-party API or the OpenRouter slug minimax/minimax-m3.