DeepSWE
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
Context window on the 2026-08-19 OpenRouter row: 262,144 tokens.
Max completion tokens on that row: 131,072.
Input / output on that row: $0.08 / $0.45 per 1M tokens.
Architecture fields: text+image->text · Gemini · gemma.
Hugging Face id on that row: google/gemma-3-27b-it.
Gemma 3 introduces multimodality, supporting vision-language input and text outputs.
When you want the 2025-03-12 catalog SKU, not a later rename.
No independent board lists this exact SKU as of 2026-08-19.
Closest benchmarked sibling: Gemini 3.7 Flash, 65% ±2% Pass@1 at high on the DeepSWE official board. That is a score for the sibling, not for this SKU.
Boards checked: DeepSWE official (2026-08-13), Artificial Analysis Intelligence Index, vals.ai hero index. Every card that does carry a row is on the coding leaderboard.
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
No Intelligence Index integer fetched for this identity. Chip omitted.
We did not open a vals.ai card for this identity. Chip omitted.
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.
Official weights id on the 2026-08-19 OpenRouter row: google/gemma-3-27b-it. Community GGUF is a quant, not a second Continuum card.
Hugging Face Hub API 2026-08-19: gated=manual.
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-00012.safetensors | 4,854,573,696 |
model-00002-of-00012.safetensors | 4,954,792,944 |
model-00003-of-00012.safetensors | 4,954,792,976 |
model-00004-of-00012.safetensors | 4,954,793,016 |
model-00005-of-00012.safetensors | 4,954,793,016 |
model-00006-of-00012.safetensors | 4,954,793,016 |
model-00007-of-00012.safetensors | 4,954,793,016 |
model-00008-of-00012.safetensors | 4,954,793,016 |
model-00009-of-00012.safetensors | 4,954,793,016 |
model-00010-of-00012.safetensors | 4,954,793,016 |
model-00011-of-00012.safetensors | 4,954,793,016 |
model-00012-of-00012.safetensors | 462,476,696 |
tokenizer.json | 33,384,568 |
tokenizer.model | 4,689,074 |
tokenizer_config.json | 1,156,999 |
model.safetensors.index.json | 126,651 |
README.md | 25,123 |
.gitattributes | 2,450 |
chat_template.json | 1,615 |
config.json | 972 |
special_tokens_map.json | 662 |
preprocessor_config.json | 570 |
generation_config.json | 215 |
processor_config.json | 70 |
added_tokens.json | 35 |
All 25 files the tree API returned are listed: every one of the 12 safetensor shards plus 13 other files. Files tab: https://huggingface.co/google/gemma-3-27b-it/tree/main.
OpenRouter slug google/gemma-3-27b-it on the 2026-08-19 catalog.
First party: https://ai.google.dev/gemini-api/docs/models.
Not on the Continuum host list fetched 2026-08-19.
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 2026-08-19: $0.08 in / $0.45 out per 1M tokens. Context 262,144 in / 131,072 out.
No separate internal-reasoning price on that row.
| Claim | Source | As of |
|---|---|---|
| OpenRouter id google/gemma-3-27b-it; context 262144; created 1741756359; $0.08 / $0.45 per 1M | OpenRouter /api/v1/models | 2026-08-19 |
| OpenRouter id google/gemma-3-27b-it, context 262,144 | OpenRouter /api/v1/models | 2026-08-19 |
| HF downloads 662,846 | Hugging Face API google/gemma-3-27b-it | 2026-08-19 |
| HF safetensors.total: 27,432,406,640 stored tensors; HF API usedStorage 179,818,970,042 bytes; created 2025-03-01T19:10:19.000Z; HF tensors BF16 | Hugging Face API google/gemma-3-27b-it | 2026-08-19 |
| 12 safetensor shards; shard bytes 54,864,980,440; tree files 25 | Hugging Face tree API google/gemma-3-27b-it | 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 Google's model family with Gemini 3.5 Flash Lite.
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 google/gemma-3-27b-it.