DeepSWE
No official DeepSWE mini-swe-agent row for this identity on the public board we fetched.
Current Upstage flagship. 497B and #29 on /models Top Weekly. Closed. Official first-party table prints two prices: standard $0.30 / $1.20, and a 90% off window at $0.03 / $0.12 through 10 Sep 2026 UTC. AA Intelligence Index 42. No official DeepSWE or vals card.
Closed API. OpenRouter hugging_face_id is empty. Not listed as a Continuum hosted id. Catalog created 10 Aug 2026 (unix 1786371636). 524,288 in / 131,072 out; text→text.
Upstage pricing/api card copy (fetched 19 Aug 2026): “Agentic LLM built for multi-step real work: tool calling, terminal tasks, and long-document reasoning.” That is vendor marketing, not a DeepSWE row.
OpenRouter supported_parameters on this slug: tools, tool_choice, structured_outputs, response_format, reasoning, include_reasoning, temperature, top_p, frequency_penalty, presence_penalty, max_tokens. Tokenizer labelled Other.
Use Solar Pro 4 when you want Upstage’s current API and you accept the promo/standard price split. AA Intelligence Index 42. We do not invent a DeepSWE percent.
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 |
|---|---|---|---|---|---|
| AA list pair | $0.30 / $1.20 per 1M | Printed AA pricing sentence. Matches official standard, not the promo. | Artificial Analysis Solar Pro 4 | source | 2026-08-19 |
| AA throughput | 42.6 tok/s | Printed Speed row. Not a coding score. | Artificial Analysis Solar Pro 4 | source | 2026-08-19 |
| AA Index eval cost | $320.92 | Printed “cost to evaluate … on the Intelligence Index.” Not $/task on DeepSWE. | Artificial Analysis Solar Pro 4 | 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.
Upstage does not publish Solar Pro 4 weights. OpenRouter hugging_face_id is empty. No Files button.
Lab card and OpenRouter listing only. No Hugging Face Files button, because there is no official repo to point at.
Fetched OpenRouter catalog on 2026-08-19. hugging_face_id was empty.
Not listed as a Continuum hosted id. First-party Upstage API or OpenRouter upstage/solar-pro4.
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.
No DeepSWE $/task: that board has no Solar Pro 4 row.
Official first-party (19 Aug 2026): standard $0.30 / $0.06 cached / $1.20 per million; 90% off through 10 Sep 2026 UTC at $0.03 / $0.006 / $0.12. OpenRouter reprints the promo input/output and the $0.006 cache read. AA reprints the standard $0.30 / $1.20.
Artificial Analysis Intelligence Index 42, fetched 19 Aug 2026. Same page prints $0.30 / $1.20, 42.6 tok/s, Speed #131 / 172, Cost #47 / 172, and $320.92 to run the Index. The AA tile is the published integer 42. AA did not print a DeepSWE-style CI, step count, or per-eval percent in the server HTML we saved. Those nine Index eval names stay unnamed here.
| Claim | Source | As of |
|---|---|---|
| Official standard $0.30 / $0.06 cached / $1.20 per 1M; promo $0.03 / $0.006 / $0.12 through 2026-09-10 UTC (90% off) | Upstage pricing/api Solar Pro 4 card + promo script v1.2 | 2026-08-19 |
| OR catalog $0.03 / $0.12, cache read $0.006/M, 524,288 / 131,072, created 1786371636, tools + reasoning, hugging_face_id empty | OpenRouter catalog | 2026-08-19 |
| AA Index 42; printed $0.30 / $1.20; 42.6 tok/s; $320.92 to evaluate Index | Artificial Analysis Solar Pro 4 | 2026-08-19 |
| vals.ai upstage_solar-pro4 404 | vals.ai Solar Pro 4 | 2026-08-19 |
| 497B week tokens, #29 Top Weekly | OpenRouter rankings This Week / Top Weekly | 2026-08-19 |
| OpenRouter id upstage/solar-pro4, context 524,288 | OpenRouter /api/v1/models | 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 Upstage's model family with Solar Pro 3.
Both are printed. Official Upstage standard is $0.30 / $1.20. A dated 90% off window (through 10 Sep 2026 UTC) is $0.03 / $0.12, and that is what the OpenRouter catalog reprints today. After the window the first-party card returns to standard.
The official DeepSWE board updated 13 Aug 2026 did not list Solar Pro 4. Omitting is the rule.
https://www.vals.ai/models/upstage_solar-pro4 returned 404 on 19 Aug 2026. Chip omitted.
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 upstage/solar-pro4.