Mercury 2
Closed weights. No supported local deployment or GPU requirement is published.
GeekRadar / The comparison desk
COMPARE.
AI modelsMercury 2
AI model Released
- API mode
- Chat*
- Context window
- 128,000 tokens
- Maximum output
- 50,000 tokens
- Open weights
- No
Capability & preference
Human preference · Arena
Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.
Recorded results
Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.
Arena text preference / chinesehuman preference · 2026-10-28 · rating
Mercury 2
1,416.86 rating
Confidence interval: 1,379.46 – 1,454.25
233 votes
- Category
- chinese
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Mercury 2
1,390.37 rating
Confidence interval: 1,370.27 – 1,410.46
801 votes
- Category
- coding
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Mercury 2
1,286.87 rating
Confidence interval: 1,261.10 – 1,312.64
542 votes
- Category
- creative_writing
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Mercury 2
1,378.44 rating
Confidence interval: 1,361.92 – 1,394.97
1202 votes
- Category
- english
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Mercury 2
1,312.39 rating
Confidence interval: 1,297.21 – 1,327.57
2238 votes
- Category
- exclude_ties
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Mercury 2
1,357.90 rating
Confidence interval: 1,322.53 – 1,393.27
253 votes
- Category
- expert
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Mercury 2
1,361.77 rating
Confidence interval: 1,348.44 – 1,375.09
1845 votes
- Category
- hard_prompts
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Mercury 2
1,373.53 rating
Confidence interval: 1,352.25 – 1,394.81
722 votes
- Category
- hard_prompts_english
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Mercury 2
1,349.33 rating
Confidence interval: 1,325.29 – 1,373.38
512 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating
Mercury 2
1,314.58 rating
Confidence interval: 1,292.60 – 1,336.55
690 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating
Mercury 2
1,344.23 rating
Confidence interval: 1,318.26 – 1,370.20
493 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating
Mercury 2
1,364.96 rating
Confidence interval: 1,327.84 – 1,402.08
232 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating
Mercury 2
1,384.84 rating
Confidence interval: 1,367.63 – 1,402.05
1112 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating
Mercury 2
1,301.23 rating
Confidence interval: 1,279.55 – 1,322.91
736 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating
Mercury 2
1,328.26 rating
Confidence interval: 1,309.14 – 1,347.39
887 votes
- Category
- instruction_following
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating
Mercury 2
1,329.63 rating
Confidence interval: 1,310.39 – 1,348.86
918 votes
- Category
- longer_query
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating
Mercury 2
1,356.74 rating
Confidence interval: 1,331.92 – 1,381.56
551 votes
- Category
- multi_turn
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating
Mercury 2
1,330.75 rating
Confidence interval: 1,317.74 – 1,343.77
1989 votes
- Category
- non_english
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating
Mercury 2
1,354.79 rating
Confidence interval: 1,344.42 – 1,365.16
3191 votes
- Category
- overall
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating
Mercury 2
1,304.09 rating
Confidence interval: 1,276.84 – 1,331.35
451 votes
- Category
- russian
- Evaluated configuration
- Mercury 2
- Evaluation type
- human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating
Mercury 2
1,416.97 rating
Confidence interval: 1,379.56 – 1,454.39
233 votes
- Category
- chinese
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
Mercury 2
1,390.17 rating
Confidence interval: 1,370.08 – 1,410.27
802 votes
- Category
- coding
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
Mercury 2
1,288.68 rating
Confidence interval: 1,263.13 – 1,314.23
547 votes
- Category
- creative_writing
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
Mercury 2
1,378.79 rating
Confidence interval: 1,362.32 – 1,395.25
1208 votes
- Category
- english
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
Mercury 2
1,312.39 rating
Confidence interval: 1,297.22 – 1,327.57
2240 votes
- Category
- exclude_ties
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
Mercury 2
1,357.92 rating
Confidence interval: 1,322.52 – 1,393.31
253 votes
- Category
- expert
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
Mercury 2
1,361.68 rating
Confidence interval: 1,348.38 – 1,374.98
1850 votes
- Category
- hard_prompts
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
Mercury 2
1,373.43 rating
Confidence interval: 1,352.24 – 1,394.63
726 votes
- Category
- hard_prompts_english
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating
Mercury 2
1,349.50 rating
Confidence interval: 1,325.44 – 1,373.55
512 votes
- Category
- industry_business_and_management_and_financial_operations
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
Mercury 2
1,315.02 rating
Confidence interval: 1,293.17 – 1,336.87
694 votes
- Category
- industry_entertainment_and_sports_and_media
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
Mercury 2
1,344.35 rating
Confidence interval: 1,318.38 – 1,370.33
493 votes
- Category
- industry_life_and_physical_and_social_science
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
Mercury 2
1,363.46 rating
Confidence interval: 1,326.23 – 1,400.68
231 votes
- Category
- industry_medicine_and_healthcare
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
Mercury 2
1,384.47 rating
Confidence interval: 1,367.26 – 1,401.68
1114 votes
- Category
- industry_software_and_it_services
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
Mercury 2
1,302.45 rating
Confidence interval: 1,280.90 – 1,324.00
741 votes
- Category
- industry_writing_and_literature_and_language
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
Mercury 2
1,327.96 rating
Confidence interval: 1,308.83 – 1,347.09
888 votes
- Category
- instruction_following
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
Mercury 2
1,329.54 rating
Confidence interval: 1,310.30 – 1,348.77
919 votes
- Category
- longer_query
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
Mercury 2
1,356.81 rating
Confidence interval: 1,331.98 – 1,381.64
551 votes
- Category
- multi_turn
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
Mercury 2
1,330.55 rating
Confidence interval: 1,317.52 – 1,343.57
1989 votes
- Category
- non_english
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
Mercury 2
1,354.86 rating
Confidence interval: 1,344.51 – 1,365.21
3197 votes
- Category
- overall
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
Mercury 2
1,304.13 rating
Confidence interval: 1,276.87 – 1,331.40
451 votes
- Category
- russian
- Evaluated configuration
- mercury-2
- Evaluation type
- human preference
Run it locally
API cost
Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.
Mercury 2
Capabilities
- Context window
- 128,000 tokens
- Maximum input
- 128,000 tokens*
- Maximum output
- 50,000 tokens
- Reasoning
- Yes
- Tool calling
- Yes
- Structured output
- Yes*
- Open weights
- No
- API mode
- Chat*
- Prompt caching
- Yes*
- System messages
- Yes*
- Tool choice
- Yes*
- Model license
- Proprietary
Input modalities
- text
Output modalities
- text
System configuration
Detailed specifications
Specifications · 13 fields
- cache read input token cost
- 2.5e-8*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- input cost per token
- 2.5e-7*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- litellm provider
- inception*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- max input tokens
- 128,000*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- max output tokens
- 50,000*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- max tokens
- 50,000*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- mode
- chat*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- output cost per token
- 7.5e-7*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports function calling
- Yes*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports prompt caching
- Yes*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports response schema
- Yes*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports system messages
- Yes*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
- supports tool choice
- Yes*
Scope and conditions
- native_context
- provider
- inception*
- mode
- chat*
- full_native_model_id
- inception/mercury-2*
- pricing_scope
- Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
0 selected Compare 0
* Provisional data