DeepSWE
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences. It integrates enhanced multimodal alignment and...
Context window on the 2026-08-19 OpenRouter row: 131,072 tokens.
Max completion tokens on that row: 32,768.
Input / output on that row: $0.18 / $2.10 per 1M tokens.
Architecture fields: text+image->text · Qwen3.
Hugging Face id on that row: Qwen/Qwen3-VL-8B-Thinking.
Qwen3-VL-8B-Thinking is the reasoning-optimized variant of the Qwen3-VL-8B multimodal model, designed for advanced visual and textual reasoning across complex scenes, documents, and temporal sequences.
When you want the 2025-10-14 catalog SKU, not a later rename.
No independent board lists this exact SKU as of 2026-08-19.
Closest benchmarked sibling: Qwen3.8 Max, 57% ±3% Pass@1 at xhigh 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: Qwen/Qwen3-VL-8B-Thinking. Community GGUF is a quant, not a second Continuum card.
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-00004.safetensors | 4,902,275,944 |
model-00002-of-00004.safetensors | 4,915,962,496 |
model-00003-of-00004.safetensors | 4,999,831,048 |
model-00004-of-00004.safetensors | 2,716,270,024 |
tokenizer.json | 7,032,403 |
vocab.json | 2,776,833 |
merges.txt | 1,671,839 |
model.safetensors.index.json | 67,759 |
tokenizer_config.json | 10,781 |
README.md | 7,201 |
chat_template.json | 5,412 |
.gitattributes | 1,519 |
config.json | 1,474 |
preprocessor_config.json | 390 |
video_preprocessor_config.json | 385 |
generation_config.json | 270 |
All 16 files the tree API returned are listed: every one of the 4 safetensor shards plus 12 other files. Files tab: https://huggingface.co/Qwen/Qwen3-VL-8B-Thinking/tree/main.
OpenRouter slug qwen/qwen3-vl-8b-thinking on the 2026-08-19 catalog.
First party: https://qwen.ai/.
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: $0.18 in / $2.10 out per 1M tokens. Context 131,072 in / 32,768 out.
No separate internal-reasoning price on that row.
| Claim | Source | As of |
|---|---|---|
| OpenRouter id qwen/qwen3-vl-8b-thinking; context 131072; created 1760463746; $0.18 / $2.10 per 1M | OpenRouter /api/v1/models | 2026-08-19 |
| OpenRouter id qwen/qwen3-vl-8b-thinking, context 131,072 | OpenRouter /api/v1/models | 2026-08-19 |
| HF downloads 91,981 | Hugging Face API Qwen/Qwen3-VL-8B-Thinking | 2026-08-19 |
| HF safetensors.total: 8,767,123,696 stored tensors; HF API usedStorage 17,534,339,512 bytes; created 2025-10-11T07:24:34.000Z; HF tensors BF16 | Hugging Face API Qwen/Qwen3-VL-8B-Thinking | 2026-08-19 |
| 4 safetensor shards; shard bytes 17,534,339,512; tree files 16 | Hugging Face tree API Qwen/Qwen3-VL-8B-Thinking | 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 Qwen's model family with Qwen3 VL 8B Instruct.
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 qwen/qwen3-vl-8b-thinking.