VerdictPal
Join

Models / Other

Solar Open2 250B

Solar · Released 2026-08-12

Upstage Solar Open 2, a 250B-A15B MoE for long-horizon agent work. Artificial Analysis Intelligence Index 37.4. AA lists $0 / $0 per 1M tokens.

Open Solar checkpoint aimed at million-token agent trajectories. Prefer sourced ledger rows over the technical-report table.

#79 of 166 on AA Index · current snapshot
37 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite47
Context1M
Input / 1MFree
Output / 1MFree
Knowledge cutoffunpublished
Statussolid

Family profile

Best published score in each covered benchmark family.

61/100 avg
Reasoning
86
Coding
46
Agentic
44
Long context
68

4 tested benchmark families

Editor's note
Recorded from the 2026-08-15 AA snapshot (index 37.4) plus arXiv 2607.20062. Editorial depth still needs a desk pass.
Benchmark placements

Where Solar Open2 250B places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
τ³-Bench BankingAgentic#54/ 10622Artificial Analysis2026-09-01
SciCodeCoding#78/ 18046Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#84/ 18029Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#92/ 18437Artificial Analysis2026-09-01
GPQA DiamondReasoning#92/ 18586Artificial Analysis2026-09-01
Terminal-BenchAgentic#100/ 18444Artificial Analysis2026-09-01
AA-LCRLong context#104/ 17968Artificial Analysis2026-09-01
AA output speedPerformance#185/ 185Artificial Analysis2026-09-01
AA time to first tokenPerformance#185/ 185Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where Solar Open2 250B shows up, it's highlighted.

ReasoningSolar Open2 250B · 86
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingSolar Open2 250B · 46
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticSolar Open2 250B · 44
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextSolar Open2 250B · 68
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Solar Open2 250B in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context68.33%Artificial Analysis2
AA output speedPerformance0 t/sArtificial Analysis2
AA time to first tokenPerformance0.00sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning37.4%Artificial Analysis2
GPQA DiamondReasoning85.7%Artificial Analysis2
Humanity's Last ExamReasoning28.5%Artificial Analysis2
SciCodeCoding45.6%Artificial Analysis2
τ³-Bench BankingAgentic21.65%Artificial Analysis2
Terminal-BenchAgentic44.19%Artificial Analysis2
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning86
Coding46
Agentic44
Long context68
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Solar Open2 250B. Scores link to the original source; gaps mean no public row exists yet.

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for Solar Open2 250B.

Reasoning

Coding

Agentic

Long context

Performance

Changelog
  1. Solar Open2 250B appears on Artificial Analysis (index 37.4, $0 / $0).
Sources