Claude 3.5 Haiku vs Solar Open2 250B
Wins 0 of 5 areas
—
Wins 5 of 5 areas
Coding · Agents · Reasoning · Facts · Long documents
Solar Open2 250B is the stronger all-rounder.
Scores updated · 14 tests both models report · How we compare
Where each one wins
Tests won in each of the five areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking03Solar Open2 250B3 of 3 tests
- CodingWriting and fixing software02Solar Open2 250B2 of 2 tests
- AgentsCarrying out multi-step tasks on its own02Solar Open2 250B2 of 2 tests
- FactsGetting facts right instead of making them up02Solar Open2 250B2 of 2 tests
- Long documentsFinding answers in very long texts01Solar Open2 250B1 of 1 test
Long documents rests on a single test.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Solar Open2 250B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+44.9points ahead
- Reasons across sets of long documentsAA-LCR+44.4points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+34.1points ahead
Where Claude 3.5 Haiku pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingSolar Open2 250B
- Terminal-Bench 2.1Solar Open2 250B by 34.110.144.2+34.1
- SciCodeSolar Open2 250B by 20.627.448+20.6
AgentsSolar Open2 250B
- GDPValSolar Open2 250B by 27.6027.6+27.6
- τ-Bench V3 · BankingSolar Open2 250B by 15.85.821.6+15.8
ReasoningSolar Open2 250B
- GPQA DiamondSolar Open2 250B by 44.940.885.7+44.9
- Humanity's Last ExamSolar Open2 250B by 24.93.628.5+24.9
- CritPtSolar Open2 250B by 5.705.7+5.7
FactsSolar Open2 250B
- AA-Omniscience · Non-hallucinationSolar Open2 250B by 15.758.974.6+15.7
- AA-Omniscience · AccuracySolar Open2 250B by 5.613.218.8+5.6
Long documentsSolar Open2 250B
- AA-LCRSolar Open2 250B by 44.427.371.7+44.4
Other results4 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- Artificial Analysis Coding IndexSolar Open2 250B by 29.115.945+29.1
- AA Agentic IndexSolar Open2 250B by 25.91.927.8+25.9
- AA-OmniscienceSolar Open2 250B by 20.7-22.5-1.8+20.7
- AA IntelligenceSolar Open2 250B by 15.88.924.7+15.8
Questions people ask
Which is better, Claude 3.5 Haiku or Solar Open2 250B?
Solar Open2 250B wins all five areas where both have results: coding, agents, reasoning, facts and long documents. Claude 3.5 Haiku wins none.
Which is better for coding?
Solar Open2 250B. It wins 2 of the 2 coding tests both models report; Claude 3.5 Haiku wins none.
How do you compare the two?
We use the 14 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 4 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.