Mercury 3 vs Qwen3.5-35B-A3B
Mercury 3 costs 3.7x less per output token than Qwen3.5-35B-A3B ($0.35 vs $1.30 per million). Updated October 23, 2026; no benchmark has tested both models yet.
- Coding
- 29.3
- Overall
- No data
- Output $/M
- $0.35
- Coding
- No data
- Overall
- 53.5
- Output $/M
- $1.30
Mercury 3 or Qwen3.5-35B-A3B: which to pick
Only Mercury 3 has a coding score, 29.3 on the Vibe Coding Index; Qwen3.5-35B-A3B has none, so the two cannot be ranked on coding.
Compare them on price, context and the boards below.
No benchmark has tested both yet. Mercury 3 has results on Terminal-Bench 4 (Vals AI), Vibe Code Bench (Vals AI) and Tax Agent Bench (Vals AI); Qwen3.5-35B-A3B on LMArena WebDev, Terminal-Bench Hard (Artificial Analysis run) and ALE-Bench. Compare them on price and context instead.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Only Mercury 3 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
No call
Only Qwen3.5-35B-A3B has a AA Intelligence Index result, 19.3.
Directional only
Web design
Design Arena: websites
No call
Only Mercury 3 has a Design Arena: websites result, 1226.
Directional only
General
Vals Index
No call
Only Mercury 3 has a Vals Index result, 43.7%.
Directional only
Long documents
Listed context window
Mercury 3
Mercury 3 takes 1M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Mercury 3
Mercury 3 $0.0003, Qwen3.5-35B-A3B $0.0008 per turn.
List prices
At a glance
- Coding index
- Mercury 3 29.3 from 2 benchmarks, Qwen3.5-35B-A3B not benchmarked.
- Widest category
- None yet: no category has been tested on both models.
- Widest single gap
- None yet: no benchmark has tested both models.
- Output price
- Mercury 3 $0.35 against Qwen3.5-35B-A3B $1.30 per million output tokens, 3.7x apart.
- Context window
- Mercury 3 takes 1M tokens, Qwen3.5-35B-A3B 262K.
Benchmark by benchmark
Every board either model has a result on, a category to a fold; bars are each result on a 0 to 100 scale. Only one model is measured on each, so no row names a leader.
CodingDirectional onlySix of these boards feed the Vibe Coding Index9 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1254 | Only Qwen3.5-35B-A3B measured |
| Terminal-Bench 4 (Vals AI) | 3.0% | Not measured | Only Mercury 3 measured |
| Vibe Code Bench (Vals AI) | 60.9% | Not measured | Only Mercury 3 measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 26.5% | Only Qwen3.5-35B-A3B measured |
| ALE-Bench | Not measured | 221.8 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: coding | Not measured | 1410 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: software and IT | Not measured | 1415 | Only Qwen3.5-35B-A3B measured |
| LMArena WebDev: React | Not measured | 1238 | Only Qwen3.5-35B-A3B measured |
| LMArena WebDev: HTML | Not measured | 1301 | Only Qwen3.5-35B-A3B measured |
| Neither model has a result on the other 34 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 19.3 | Only Qwen3.5-35B-A3B measured |
| Epoch Capabilities Index | Not measured | 142.5 | Only Qwen3.5-35B-A3B measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index5 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| Tax Agent Bench (Vals AI) | 18.2% | Not measured | Only Mercury 3 measured |
| Legal Research Bench (Vals AI) | 28.9% | Not measured | Only Mercury 3 measured |
| Harvey's Legal Agent Benchmark (Vals AI) | 5.0% | Not measured | Only Mercury 3 measured |
| τ²-Bench Telecom | Not measured | 89.2% | Only Qwen3.5-35B-A3B measured |
| Vending-Bench 2 | Not measured | $463 | Only Qwen3.5-35B-A3B measured |
| Neither model has a result on the other 37 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 84.5% | Only Qwen3.5-35B-A3B measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 21.0% | Only Qwen3.5-35B-A3B measured |
| GPQA Diamond (Epoch AI run) | Not measured | 83.5% | Only Qwen3.5-35B-A3B measured |
| LMArena Text: hard prompts | Not measured | 1400 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: expert | Not measured | 1407 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: hard prompts in English | Not measured | 1411 | Only Qwen3.5-35B-A3B measured |
| Neither model has a result on the other 17 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| OTIS Mock AIME | Not measured | 70.0% | Only Qwen3.5-35B-A3B measured |
| LMArena Text: math | Not measured | 1404 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: mathematical work | Not measured | 1410 | Only Qwen3.5-35B-A3B measured |
| MathArena | Not measured | 14.7% | Only Qwen3.5-35B-A3B measured |
| Neither model has a result on the other 10 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index4 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| LMArena Text: science | Not measured | 1413 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: medicine and healthcare | Not measured | 1421 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: legal and government | Not measured | 1398 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: business and finance | Not measured | 1394 | Only Qwen3.5-35B-A3B measured |
| Neither model has a result on the other 16 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| Design Arena: websites | 1226 | Not measured | Only Mercury 3 measured |
| Design Arena: UI components | 1247 | Not measured | Only Mercury 3 measured |
| Design Arena: game dev | 1227 | Not measured | Only Mercury 3 measured |
| Design Arena: data viz | 1201 | Not measured | Only Mercury 3 measured |
| Design Arena: all code categories | 1231 | Not measured | Only Mercury 3 measured |
| Design Arena: SVG | 1174 | Not measured | Only Mercury 3 measured |
| Neither model has a result on the other 3 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index22 boards measured
| Benchmark | Mercury 3 | Qwen3.5-35B-A3B | Leader |
|---|---|---|---|
| Vals Index | 43.7% | Not measured | Only Mercury 3 measured |
| IFBench | Not measured | 72.5% | Only Qwen3.5-35B-A3B measured |
| Long context reasoning (AA-LCR) | Not measured | 72.0% | Only Qwen3.5-35B-A3B measured |
| DTBench | Not measured | 80.0% | Only Qwen3.5-35B-A3B measured |
| LMCA | Not measured | 29.5% | Only Qwen3.5-35B-A3B measured |
| LMArena Text | Not measured | 1395 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: instruction following | Not measured | 1379 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: creative writing | Not measured | 1347 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: English | Not measured | 1407 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: non-English | Not measured | 1378 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: Chinese | Not measured | 1456 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: German | Not measured | 1367 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: Japanese | Not measured | 1326 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: Korean | Not measured | 1354 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: French | Not measured | 1413 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: Spanish | Not measured | 1390 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: Russian | Not measured | 1377 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: Polish | Not measured | 1385 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: multi-turn | Not measured | 1390 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: longer queries | Not measured | 1389 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: writing and language | Not measured | 1365 | Only Qwen3.5-35B-A3B measured |
| LMArena Text: entertainment and media | Not measured | 1347 | Only Qwen3.5-35B-A3B measured |
| Neither model has a result on the other 26 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Mercury 3 costs less
Repository review
50K input + 3K output tokens
Mercury 3 costs less
Agent loop
200K input + 10K output tokens
Mercury 3 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Inception Mercury 3 | Qwen3.5-35B-A3B |
|---|---|---|
| Maker | Inception | Qwen |
| Weights | No public weights listed | Open weights |
| Released | October 16, 2026 | February 25, 2026 |
| Context window | 1M | 262K |
| Input $/Mtok | $0.10 | $0.1625 |
| Output $/Mtok | $0.35 | $1.30 |
| Coding index | 29.3 | Not benchmarked |
| Independent measurements behind it | 2 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Mercury 3 better than Qwen3.5-35B-A3B for coding?
Only Mercury 3 has a coding score, 29.3 on the Vibe Coding Index; Qwen3.5-35B-A3B has none, so the two cannot be ranked on coding. Compare them on price, context and the boards below.
Which is cheaper, Mercury 3 or Qwen3.5-35B-A3B?
Mercury 3 costs $0.10 per million input tokens and $0.35 per million output tokens; Qwen3.5-35B-A3B costs $0.1625 and $1.30. Cost per solved task is known for neither of them.
Which has the bigger context window, Mercury 3 or Qwen3.5-35B-A3B?
Mercury 3, with 1M tokens against 262K.
Has any benchmark tested both Mercury 3 and Qwen3.5-35B-A3B?
No benchmark has tested both yet. Mercury 3 has results on Terminal-Bench 4 (Vals AI), Vibe Code Bench (Vals AI) and Tax Agent Bench (Vals AI); Qwen3.5-35B-A3B on LMArena WebDev, Terminal-Bench Hard (Artificial Analysis run) and ALE-Bench. Compare them on price and context instead. Data as of October 23, 2026.