Solar Mini 4
Upstage · released Sep 23, 2026
Solar Mini 4 is Upstage's compact, cost-efficient language model, a 35B-parameter mixture-of-experts with 3B active parameters and a 524K context window. It is built for agentic use cases where response...
Specification
- Context window
- 524K
- Max output
- 131K
- Knowledge cutoff
- Not stated
- Parameters
- Undisclosed
- Licence
- Proprietary
- Serving providers
- 1
- Moderated
- No
- Uptime
- 100.0%
Intelligence
24.1
74th percentile
Coding
—
Coding Index
Agentic
—
Agentic Index
Output speed
55 t/s
Median across providers
Latency
1.96s
Time to first token
Cost per task
$0.02
Estimated
Benchmarks
Where the score comes from
The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.
Evaluation scores
Percentage correct · higher is better
- AA-LCR (long context)83.3%
- SciCode47.6%
- Humanity's Last Exam25.8%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| AA-LCR (long context) | 83.3% |
| SciCode | 47.6% |
| Humanity's Last Exam | 25.8% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- Claude Sonnet 4.6Anthropic24.7
- Ling 3.0 Flash VLInclusionAI24.6
- Qwen3 MaxQwen24.5
- Ling 3.0 Tiny (free)InclusionAI24.3
- Solar Mini 4Upstage24.1
- GPT-5.4 MiniOpenAI24.1
- Claude Opus 4.5Anthropic23.7
- GPT-5.1-CodexOpenAI23.7
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| Claude Sonnet 4.6 | 24.7 |
| Ling 3.0 Flash VL | 24.6 |
| Qwen3 Max | 24.5 |
| Ling 3.0 Tiny (free) | 24.3 |
| Solar Mini 4 | 24.1 |
| GPT-5.4 Mini | 24.1 |
| Claude Opus 4.5 | 23.7 |
| GPT-5.1-Codex | 23.7 |
Percentile among all indexed models
Pricing
What it costs to run
List prices per million tokens, plus what one representative task works out to.
List price
- Input / 1M tokens
- $0.05
- Output / 1M tokens
- $0.2
- Cached input / 1M
- $0.005
- Blended 3:1
- $0.088
One task, estimated
$0.02
- Input tokens
- 50,000
- Output tokens
- 80,000
- Profile
- Reasoning
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).
Serving providers
Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.
Arena
Head-to-head generation quality
Elo from pairwise judgements, broken out by the kind of thing the model was asked to build.
- Overall Elo
- 1,160
- Win rate
- 33.1%
- Strongest at
- Websites
- Tournaments
- 1,095
Elo by category
Dot position on a 1,125–1,175 scale · Elo has no meaningful zero
- 1,160
- 1,144
Agent categories (full-stack apps, mobile apps) are only scored for models tested in agent harnesses.
View as table
| Category | Elo | Win rate |
|---|---|---|
| Websites | 1160 | 33.1% |
| Game development | 1144 | 30.4% |