Upstage model product
A reasoning-focused agentic text-to-text LLM from Upstage for multi-step tasks.
Updated Sep 8, 2026. Default version: Solar Pro 4
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Solar Pro 4.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkTau3 Banking | Score23.00% | Rank03 | Participants7 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkAA-LCR | Score71.00% | Rank04 | Participants19 | Percentile83.33% | EvidenceC | Evaluated |
| BenchmarkLiveCodeBench | Score87.80% | Rank04 | Participants75 | Percentile95.95% | EvidenceC | Evaluated |
| BenchmarkAIME 2026 | Score95.30% | Rank07 | Participants22 | Percentile71.43% | EvidenceC | Evaluated |
| BenchmarkMMLU-Pro | Score86.30% | Rank13 | Participants138 | Percentile91.24% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.1 | Score57.00% | Rank29 | Participants35 | Percentile17.65% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score-1.32% | Rank31 | Participants49 | Percentile37.50% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score89.00% | Rank33 | Participants247 | Percentile86.99% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score-8.60% | Rank42 | Participants49 | Percentile14.58% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score-9.87% | Rank43 | Participants49 | Percentile12.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score-0.79% | Rank43 | Participants49 | Percentile12.50% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score-13.71% | Rank45 | Participants49 | Percentile8.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Recovery Steps | Score-24.95% | Rank46 | Participants49 | Percentile6.25% | EvidenceA | Evaluated |
| BenchmarkBrowseComp | Score49.20% | Rank49 | Participants63 | Percentile22.58% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score70.60% | Rank62 | Participants113 | Percentile45.54% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,370.72 rating | Rank68 | Participants97 | Percentile30.21% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,396.98 rating | Rank105 | Participants121 | Percentile13.33% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,377.54 rating | Rank115 | Participants210 | Percentile45.45% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,376.84 rating | Rank127 | Participants210 | Percentile39.71% | EvidenceA | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent steerability | Categoryoverall | Rank31 | Rating / score-0.01 | VotesN/A | Observations3.4K | Result date |
| Arenaagent task outcome explicit | Categoryoverall | Rank42 | Rating / score-0.09 | VotesN/A | Observations2.8K | Result date |
| Arenaagent | Categoryoverall | Rank43 | Rating / score-0.10 | VotesN/A | Observations199K | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank43 | Rating / score-0.01 | VotesN/A | Observations181.6K | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank45 | Rating / score-0.14 | VotesN/A | Observations970 | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank46 | Rating / score-0.25 | VotesN/A | Observations10.4K | Result date |
| Arenawebdev | Categorywebdev-react | Rank64 | Rating / score1,356.83 | Votes931 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev-html | Rank65 | Rating / score1,394.34 | Votes159 | ObservationsN/A | Result date |
| Arenawebdev | Categoryoverall | Rank68 | Rating / score1,370.72 | Votes1,404 | ObservationsN/A | Result date |
| Arenawebdev | Categorywebdev | Rank68 | Rating / score1,370.72 | Votes1,404 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts english | Rank79 | Rating / score1,454.69 | Votes649 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank90 | Rating / score1,425.09 | Votes653 | ObservationsN/A | Result date |
| Arenatext factuality | Categorycoding | Rank90 | Rating / score1,459.26 | Votes684 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank92 | Rating / score1,417.81 | Votes491 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryenglish | Rank93 | Rating / score1,426.42 | Votes994 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry software and it services | Rank93 | Rating / score1,449.09 | Votes957 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank95 | Rating / score1,425.04 | Votes684 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryinstruction following | Rank97 | Rating / score1,393.38 | Votes879 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank98 | Rating / score1,418.31 | Votes957 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry entertainment and sports and media | Rank98 | Rating / score1,344.53 | Votes640 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryhard prompts | Rank100 | Rating / score1,427.36 | Votes1,623 | ObservationsN/A | Result date |
| Arenatext factuality | Categorylonger query | Rank100 | Rating / score1,410.17 | Votes1,172 | ObservationsN/A | Result date |
| Arenatext style control | Categoryhard prompts english | Rank100 | Rating / score1,437.99 | Votes653 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry writing and literature and language | Rank102 | Rating / score1,369.20 | Votes657 | ObservationsN/A | Result date |
| Arenatext | Categoryexpert | Rank104 | Rating / score1,399.01 | Votes262 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexclude ties | Rank105 | Rating / score1,379.32 | Votes1,811 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank105 | Rating / score1,396.98 | Votes2,437 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank106 | Rating / score1,405.10 | Votes994 | ObservationsN/A | Result date |
| Arenatext factuality | Categorynon english | Rank106 | Rating / score1,376.34 | Votes1,443 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank107 | Rating / score1,447.48 | Votes684 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNanoGPT | Provider model IDupstage/solar-pro4 | Regionglobal | Input / 1M$0.03 | Output / 1M$0.12 | Context524.3K | Updated |
| ProviderOpenRouter | Provider model IDupstage/solar-pro4 | Regionglobal | Input / 1M$0.03 | Output / 1M$0.12 | Context524.3K | Updated |
| ProviderLLMTR | Provider model IDupstage/solar-pro4 | Regionglobal | Input / 1M$0.03 | Output / 1M$0.12 | Context524.3K | Updated |
| ProviderUpstage | Provider model IDsolar-pro4 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context524.3K | Updated |
| ProviderKilo Gateway | Provider model IDupstage/solar-pro4 | Regionglobal | Input / 1M$0.30 | Output / 1M$1.2 | Context524.3K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Solar Pro 4. Runtime does not affect the capability score.
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionSolar Pro 4 | Released | LLMBoard56.56 | ParametersN/A | Context524.3K | Max output131.1K | Open weightsNo | LicenseProprietary |
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Solar Pro 4 and its available data.
Solar Pro 4 is Upstage's proprietary reasoning-focused agentic LLM for text-to-text tasks, supporting English, Korean, and Japanese. It has a ~512K context window, up to 128K output tokens, and adjustable reasoning effort.
Data as of 2026-09-08.
Common questions about Solar Pro 4.
Solar Pro 4's default version was released on Aug 6, 2026.
Solar Pro 4's official API price is $0.30 per million input tokens and $1.2 per million output tokens via Upstage. The lowest tracked third-party offer starts at $0.03 input and $0.12 output via NanoGPT.
Solar Pro 4 was created by Upstage.
The default version has a 524.3K token context window.
No. The default version is not marked as having publicly available weights.
5 provider offerings are linked to the default version.