DeepSWE
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
OpenRouter left hugging_face_id blank. Official weights are still the Mistral org repo above. Medium 3.5 is a sibling, not this card.
HF API (fetched 19 Aug 2026): Apache-2.0, 272 consolidated-*-of-00272.safetensors shards, usedStorage 681,538,639,639 bytes. safetensors.total was absent. Raw config.json 404s (not 401). Official README.md is public (200). Hub base_model is mistralai/Mistral-Large-3-675B-Base-2512. We will not invent a tensor count the Hub omitted.
Vendor README (raw 200, 19 Aug 2026) prints two activated-parameter claims on the same page: intro “41B active” and a bullet “673B params and 39B active”. Those are vendor copy. Hub safetensors.total is still absent. Do not write 675B as a Hub tensor count.
Use Large 3 when you want Mistral’s capable Large checkpoint. AA Intelligence Index 16. No DeepSWE row.
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
We did not open a vals.ai card for this identity. Chip omitted.
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 | $0.50 / $1.50 per 1M | Catalog row. No DeepSWE $/task: that board has no Mistral Large 3 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.
Do not invent a safetensors.total from the 675B name. The Hub did not publish one for this repo. Raw config.json 404s; README.md is public (200).
Vendor README (200) states 673B params / 39B active in one bullet and 41B active in the intro. Public card HTML also shows the 675B repo id. All vendor/UI claims. safetensors.total remains absent.
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 |
|---|---|
consolidated-00001-of-00272.safetensors | 2,544,898,528 |
consolidated-00002-of-00272.safetensors | 2,495,933,768 |
consolidated-00003-of-00272.safetensors | 2,495,933,656 |
consolidated-00004-of-00272.safetensors | 2,495,933,648 |
consolidated-00005-of-00272.safetensors | 2,495,933,656 |
consolidated-00006-of-00272.safetensors | 2,494,520,904 |
consolidated-00007-of-00272.safetensors | 2,495,933,704 |
consolidated-00008-of-00272.safetensors | 2,495,933,656 |
consolidated-00009-of-00272.safetensors | 2,495,933,648 |
consolidated-00010-of-00272.safetensors | 2,494,520,536 |
consolidated-00011-of-00272.safetensors | 2,495,933,816 |
consolidated-00012-of-00272.safetensors | 2,495,933,648 |
Showing the first 12 of 272 safetensor shards. The tree also holds 11 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 260 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 mistralai/mistral-large-2512.
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: $0.50 / $1.50 per million. No DeepSWE $/task.
Artificial Analysis Intelligence Index 16, 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 |
|---|---|---|
| OR list $0.50 / $1.50 | OpenRouter catalog | 2026-08-19 |
| HF usedStorage 681,538,639,639; 272 consolidated shards; likes 245; created 2025-11-28T18:05:12Z; safetensors.total absent | Hugging Face API Mistral-Large-3-675B-Instruct-2512 | 2026-08-19 |
| Raw config.json 404; README.md 200. Vendor README: 673B params / 39B active (bullet) and 41B active (intro). Not Hub tensors. | HF README Mistral-Large-3-675B-Instruct-2512 | 2026-08-19 |
| OpenRouter id mistralai/mistral-large-2512, context 262,144 | OpenRouter /api/v1/models | 2026-08-19 |
| HF downloads 1,643 | Hugging Face API mistralai/Mistral-Large-3-675B-Instruct-2512 | 2026-08-19 |
| 272 safetensor shards; shard bytes 681,504,807,752; tree files 283 | Hugging Face tree API mistralai/Mistral-Large-3-675B-Instruct-2512 | 2026-08-19 |
| Spaces API returned 50 (capped at 50) | 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 Mistral's model family with Mistral Medium 3.5.
Medium 3.5 is newer and 30 on the Intelligence Index. Large 3 is the capable 675B flagship. We do not swap the card to chase a mid-tier integer.
The official DeepSWE board updated 13 Aug 2026 did not list Mistral Large 3. 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 mistralai/mistral-large-2512.