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: 131,072 tokens.
Max completion tokens on that row: 131,072.
Input / output on that row: $0.03 / $0.17 per 1M tokens.
Architecture fields: text->text · GPT.
Hugging Face id on that row: openai/gpt-oss-120b.
gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.
When you want the 2025-08-05 catalog SKU, not a later rename.
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 |
|---|---|---|---|---|---|
| GDPval-AA v2 | 15% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| τ³-Banking | 12.8% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| Terminal-Bench v2.1 | 26.2% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| SciCode | 38.9% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| Humanity's Last Exam | 19.6% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| GPQA Diamond | 78.2% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| CritPt | 1.1% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| AA-Omniscience Accuracy | 21.8% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | source | 2026-08-19 |
| AA-LCR | 51% | Printed on the AA Intelligence Evaluations grid for gpt-oss-120b (high). Not a DeepSWE chip. Tile effort: high. | Artificial Analysis gpt-oss-120b (high) | 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.
Official weights id on the 2026-08-19 OpenRouter row: openai/gpt-oss-120b. 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-00000-of-00014.safetensors | 4,625,017,896 |
model-00001-of-00014.safetensors | 4,115,586,736 |
model-00002-of-00014.safetensors | 4,625,017,888 |
model-00003-of-00014.safetensors | 4,115,586,752 |
model-00004-of-00014.safetensors | 4,625,017,896 |
model-00005-of-00014.safetensors | 4,115,586,696 |
model-00006-of-00014.safetensors | 4,625,017,856 |
model-00007-of-00014.safetensors | 4,060,267,176 |
model-00008-of-00014.safetensors | 4,625,017,896 |
model-00009-of-00014.safetensors | 4,170,906,304 |
model-00010-of-00014.safetensors | 4,625,017,896 |
model-00011-of-00014.safetensors | 4,115,586,752 |
model-00012-of-00014.safetensors | 4,064,660,808 |
model-00013-of-00014.safetensors | 4,625,017,896 |
model-00014-of-00014.safetensors | 4,115,586,736 |
original/model--00001-of-00007.safetensors | 10,544,040,680 |
original/model--00002-of-00007.safetensors | 10,488,721,680 |
original/model--00003-of-00007.safetensors | 10,488,721,688 |
original/model--00004-of-00007.safetensors | 10,488,721,672 |
original/model--00005-of-00007.safetensors | 10,488,721,680 |
original/model--00006-of-00007.safetensors | 10,433,402,600 |
original/model--00007-of-00007.safetensors | 2,316,539,800 |
metal/model.bin | 65,238,253,568 |
tokenizer.json | 27,868,174 |
model.safetensors.index.json | 54,511 |
original/model.safetensors.index.json | 37,796 |
original/dtypes.json | 19,658 |
chat_template.jinja | 16,738 |
LICENSE | 11,357 |
README.md | 7,111 |
tokenizer_config.json | 4,200 |
config.json | 2,089 |
.gitattributes | 1,570 |
original/config.json | 377 |
USAGE_POLICY | 201 |
generation_config.json | 177 |
special_tokens_map.json | 98 |
All 37 files the tree API returned are listed: every one of the 22 safetensor shards plus 15 other files. Files tab: https://huggingface.co/openai/gpt-oss-120b/tree/main.
OpenRouter slug openai/gpt-oss-120b on the 2026-08-19 catalog.
First party: https://platform.openai.com/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: $0.03 in / $0.17 out per 1M tokens. Context 131,072 in / 131,072 out.
No separate internal-reasoning price on that row.
| Claim | Source | As of |
|---|---|---|
| OpenRouter id openai/gpt-oss-120b; context 131072; created 1754414231; $0.03 / $0.17 per 1M | OpenRouter /api/v1/models | 2026-08-19 |
| AA Intelligence Index 24 (high); 9 printed Index benches | Artificial Analysis model page | 2026-08-19 |
| OpenRouter id openai/gpt-oss-120b, context 131,072 | OpenRouter /api/v1/models | 2026-08-19 |
| HF downloads 4,657,776 | Hugging Face API openai/gpt-oss-120b | 2026-08-19 |
| HF safetensors.total: 116,829,156,672 stored tensors; HF API usedStorage 195,851,052,807 bytes; created 2025-08-04T22:33:06.000Z; HF tensors BF16 + U8 | Hugging Face API openai/gpt-oss-120b | 2026-08-19 |
| 22 safetensor shards; shard bytes 130,497,762,984; tree files 37 | Hugging Face tree API openai/gpt-oss-120b | 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 OpenAI's model family with gpt-oss-20b.
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 openai/gpt-oss-120b.