Claude 3.7 Sonnet vs JT-35B-Flash
Wins 2 of 6 areas
Facts · Following instructions
Wins 2 of 6 areas
Agents · Long documents
The two are evenly matched.Claude 3.7 Sonnet is better at facts and following instructions; JT-35B-Flash at agents and long documents.
Scores updated · 15 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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.
- FactsGetting facts right instead of making them up20Claude 3.7 Sonnet2 of 2 tests
- Following instructionsDoing exactly what it is asked10Claude 3.7 Sonnet1 of 1 test
- AgentsCarrying out multi-step tasks on its own01JT-35B-Flash1 of 1 test
- Long documentsFinding answers in very long texts01JT-35B-Flash1 of 1 test
- CodingWriting and fixing software11Even1 each
- ReasoningHard problems that need careful thinking11Even1 each · 1 tie
Agents, long documents and following instructions rest on a single test each.
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 three tests each model wins by the widest margin. Scores are out of 100.
Where Claude 3.7 Sonnet pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+23.8points ahead
- Code for real scientific research problemsSciCode+11.2points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+6.3points ahead
Where JT-35B-Flash pulls ahead
- Reasons across sets of long documentsAA-LCR+14points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+7.6points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+5.7points ahead
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SciCodeClaude 3.7 Sonnet by 11.240.329.1+11.2
- Terminal-Bench HardJT-35B-Flash by 7.621.228.8+7.6
ReasoningEven
- GPQA DiamondJT-35B-Flash by 5.777.282.9+5.7
- Humanity's Last ExamClaude 3.7 Sonnet by 3.39.76.4+3.3
- CritPttie0.90tie
FactsClaude 3.7 Sonnet
- AA-Omniscience · Non-hallucinationClaude 3.7 Sonnet by 23.86036.2+23.8
- AA-Omniscience · AccuracyClaude 3.7 Sonnet by 3.82824.2+3.8
Following instructionsClaude 3.7 Sonnet
- IFBenchClaude 3.7 Sonnet by 6.348.342+6.3
Other results5 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.
- τ²-Bench Telecom (AA run)JT-35B-Flash by 44.454.799.1+44.4
- AA-OmniscienceClaude 3.7 Sonnet by 23.5-0.7-24.2+23.5
- AA Agentic IndexJT-35B-Flash by 15.33752.3+15.3
- Artificial Analysis Coding IndexClaude 3.7 Sonnet by 7.536.428.9+7.5
- AA IntelligenceJT-35B-Flash by 117.718.7+1
Questions people ask
Which is better, Claude 3.7 Sonnet or JT-35B-Flash?
Claude 3.7 Sonnet and JT-35B-Flash each win two of the six areas where both have results. Claude 3.7 Sonnet wins facts and following instructions; JT-35B-Flash wins agents and long documents. They are level on coding and reasoning.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
How do you compare the two?
We use the 15 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 5 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.