1gpt-5.6 sol maxLeader
GPT-5.6 Sol · max
Raw avg
89.7
P10 floor
80.3
Cost / call
$0.37
Calls
n=50
salesevals.com/Evaluated Jul 9, 2026
Which models know sales?
48 model configurations coach GPT- and Sonnet-generated synthetic sales calls with hidden ground truth. A judge scores each coaching note from 0–100 on whether it found the real strengths, flaws, and next moves.
- Calls
- 50
- Models
- 48
- Evaluations
- 2400
- Benchmark
- 87.4
50 calls · 2400 evaluationsRank: Sales coaching benchmarkAll available runsBuild-time static dataEvals completed Jul 9, 2026
Ranked by benchmark score
Leaderboard
Weighted for needle recall, sales instinct, prioritization, technical accuracy, false-positive control, and raw score across 2400 evaluations.
90.0
2gpt-5.6 terra max
GPT-5.6 Terra · max
Raw avg
89.6
P10 floor
80.6
Cost / call
$0.10
Calls
n=50
3gpt-5.6 luna max
GPT-5.6 Luna · max
Raw avg
89.7
P10 floor
83.7
Cost / call
$0.06
Calls
n=50
4gpt-5.6 luna xhigh
GPT-5.6 Luna · xhigh
Raw avg
89.7
P10 floor
81.7
Cost / call
$0.04
Calls
n=50
5gpt-5.6 terra low
GPT-5.6 Terra · low
Raw avg
89.5
P10 floor
84.5
Cost / call
$0.05
Calls
n=50
6gpt-5.6 sol low
GPT-5.6 Sol · low
Raw avg
89.5
P10 floor
81.3
Cost / call
$0.12
Calls
n=50
7gpt-5.6 sol xhigh
GPT-5.6 Sol · xhigh
Raw avg
89.5
P10 floor
83.1
Cost / call
$0.17
Calls
n=50
8gpt-5.6 terra xhigh
GPT-5.6 Terra · xhigh
Raw avg
89.2
P10 floor
78.6
Cost / call
$0.08
Calls
n=50
9gpt-5.6 terra high
GPT-5.6 Terra · high
Raw avg
89.3
P10 floor
82.0
Cost / call
$0.06
Calls
n=50
10gpt-5.6 sol none
GPT-5.6 Sol · none
Raw avg
89.3
P10 floor
81.4
Cost / call
$0.12
Calls
n=50
11gpt-5.6 sol high
GPT-5.6 Sol · high
Raw avg
89.2
P10 floor
79.8
Cost / call
$0.12
Calls
n=50
12gpt-5.4 xhigh
GPT-5.4 · xhigh
Raw avg
89.0
P10 floor
79.9
Cost / call
$0.28
Calls
n=50
13gpt-5.6 sol medium
GPT-5.6 Sol · medium
Raw avg
89.0
P10 floor
79.7
Cost / call
$0.13
Calls
n=50
14gpt-5.4 high
GPT-5.4 · high
Raw avg
89.0
P10 floor
81.9
Cost / call
$0.10
Calls
n=50
15gpt-5.5 medium
GPT-5.5 · medium
Raw avg
88.8
P10 floor
82.0
Cost / call
$0.12
Calls
n=50
16gpt-5.5 xhigh
GPT-5.5 · xhigh
Raw avg
89.0
P10 floor
78.5
Cost / call
$0.20
Calls
n=50
17gpt-5.6 terra none
GPT-5.6 Terra · none
Raw avg
88.8
P10 floor
79.7
Cost / call
$0.05
Calls
n=50
18gpt-5.6 terra medium
GPT-5.6 Terra · medium
Raw avg
88.9
P10 floor
80.1
Cost / call
$0.05
Calls
n=50
19gpt-5.5 high
GPT-5.5 · high
Raw avg
88.6
P10 floor
79.4
Cost / call
$0.14
Calls
n=50
20gpt-5.6 luna medium
GPT-5.6 Luna · medium
Raw avg
88.6
P10 floor
82.0
Cost / call
$0.02
Calls
n=50
21gpt-5.6 luna high
GPT-5.6 Luna · high
Raw avg
88.6
P10 floor
78.9
Cost / call
$0.03
Calls
n=50
22gpt-5.6 luna low
GPT-5.6 Luna · low
Raw avg
88.5
P10 floor
79.4
Cost / call
$0.02
Calls
n=50
23gpt-5.4 medium
GPT-5.4 · medium
Raw avg
88.3
P10 floor
80.1
Cost / call
$0.06
Calls
n=50
24gpt-5.5 none
GPT-5.5 · none
Raw avg
88.1
P10 floor
76.6
Cost / call
$0.13
Calls
n=50
25gpt-5.6 luna none
GPT-5.6 Luna · none
Raw avg
87.7
P10 floor
79.7
Cost / call
$0.02
Calls
n=50
26gpt-5.5 low
GPT-5.5 · low
Raw avg
87.7
P10 floor
78.3
Cost / call
$0.14
Calls
n=50
27fable 5 high
Claude Fable 5 · high
Raw avg
87.5
P10 floor
77.0
Cost / call
$0.46
Calls
n=50
28gpt-5.4 low
GPT-5.4 · low
Raw avg
87.4
P10 floor
78.3
Cost / call
$0.05
Calls
n=50
29gpt-5.4 none
GPT-5.4 · none
Raw avg
87.4
P10 floor
78.2
Cost / call
$0.05
Calls
n=50
30opus 4.7 max
Claude Opus 4.7 · max
Raw avg
87.3
P10 floor
75.0
Cost / call
$0.21
Calls
n=50
31opus 4.7 high
Claude Opus 4.7 · high
Raw avg
86.8
P10 floor
77.8
Cost / call
$0.17
Calls
n=50
32muse spark 1.1 high
Muse Spark 1.1 · high
Raw avg
86.4
P10 floor
77.0
Cost / call
$0.03
Calls
n=50
33muse spark 1.1 medium
Muse Spark 1.1 · medium
Raw avg
86.2
P10 floor
78.4
Cost / call
$0.02
Calls
n=50
34muse spark 1.1 minimal
Muse Spark 1.1 · minimal
Raw avg
85.7
P10 floor
75.8
Cost / call
$0.02
Calls
n=50
35muse spark 1.1 low
Muse Spark 1.1 · low
Raw avg
85.4
P10 floor
75.1
Cost / call
$0.02
Calls
n=50
36opus 4.8 medium
Claude Opus 4.8 · medium
Raw avg
85.8
P10 floor
72.1
Cost / call
$0.14
Calls
n=50
37opus 4.7 medium
Claude Opus 4.7 · medium
Raw avg
85.6
P10 floor
74.0
Cost / call
$0.33
Calls
n=50
38opus 4.7 xhigh
Claude Opus 4.7 · xhigh
Raw avg
85.6
P10 floor
73.1
Cost / call
$0.17
Calls
n=50
39opus 4.7 low
Claude Opus 4.7 · low
Raw avg
85.6
P10 floor
75.6
Cost / call
$0.13
Calls
n=50
40opus 4.8 max
Claude Opus 4.8 · max
Raw avg
85.4
P10 floor
72.9
Cost / call
$0.18
Calls
n=50
41opus 4.8 xhigh
Claude Opus 4.8 · xhigh
Raw avg
85.2
P10 floor
73.0
Cost / call
$0.16
Calls
n=50
42opus 4.8 high
Claude Opus 4.8 · high
Raw avg
84.9
P10 floor
68.8
Cost / call
$0.15
Calls
n=50
43sonnet 4.6
Claude Sonnet 4.6 · default
Raw avg
84.6
P10 floor
71.9
Cost / call
$0.10
Calls
n=50
44sonnet 5
Claude Sonnet 5 · default
Raw avg
84.6
P10 floor
71.6
Cost / call
$0.05
Calls
n=50
45opus 4.8 low
Claude Opus 4.8 · low
Raw avg
84.0
P10 floor
67.2
Cost / call
$0.12
Calls
n=50
46glm 5.2
GLM 5.2 · default
Raw avg
84.0
P10 floor
71.3
Cost / call
$0.03
Calls
n=50
47deepseek v4 pro
DeepSeek V4 Pro · default
Raw avg
83.5
P10 floor
69.1
Cost / call
$0.0047
Calls
n=50
48gemini 3.1 pro previewTrailing
Gemini 3.1 Pro Preview · default
Raw avg
78.9
P10 floor
66.6
Cost / call
$0.03
Calls
n=50
Filtered benchmark87.4
| # | Model | Benchmark | Raw avg | P10 floor | Cost / call | Distribution | n |
|---|---|---|---|---|---|---|---|
| 1 | gpt-5.6 sol maxLeader GPT-5.6 Sol · max | 90.0 | 89.7 | 80.3 | $0.37 | 50 | |
| 2 | gpt-5.6 terra max GPT-5.6 Terra · max | 90.0 | 89.6 | 80.6 | $0.10 | 50 | |
| 3 | gpt-5.6 luna max GPT-5.6 Luna · max | 89.9 | 89.7 | 83.7 | $0.06 | 50 | |
| 4 | gpt-5.6 luna xhigh GPT-5.6 Luna · xhigh | 89.9 | 89.7 | 81.7 | $0.04 | 50 | |
| 5 | gpt-5.6 terra low GPT-5.6 Terra · low | 89.9 | 89.5 | 84.5 | $0.05 | 50 | |
| 6 | gpt-5.6 sol low GPT-5.6 Sol · low | 89.8 | 89.5 | 81.3 | $0.12 | 50 | |
| 7 | gpt-5.6 sol xhigh GPT-5.6 Sol · xhigh | 89.7 | 89.5 | 83.1 | $0.17 | 50 | |
| 8 | gpt-5.6 terra xhigh GPT-5.6 Terra · xhigh | 89.7 | 89.2 | 78.6 | $0.08 | 50 | |
| 9 | gpt-5.6 terra high GPT-5.6 Terra · high | 89.7 | 89.3 | 82.0 | $0.06 | 50 | |
| 10 | gpt-5.6 sol none GPT-5.6 Sol · none | 89.6 | 89.3 | 81.4 | $0.12 | 50 | |
| 11 | gpt-5.6 sol high GPT-5.6 Sol · high | 89.4 | 89.2 | 79.8 | $0.12 | 50 | |
| 12 | gpt-5.4 xhigh GPT-5.4 · xhigh | 89.3 | 89.0 | 79.9 | $0.28 | 50 | |
| 13 | gpt-5.6 sol medium GPT-5.6 Sol · medium | 89.3 | 89.0 | 79.7 | $0.13 | 50 | |
| 14 | gpt-5.4 high GPT-5.4 · high | 89.2 | 89.0 | 81.9 | $0.10 | 50 | |
| 15 | gpt-5.5 medium GPT-5.5 · medium | 89.1 | 88.8 | 82.0 | $0.12 | 50 | |
| 16 | gpt-5.5 xhigh GPT-5.5 · xhigh | 89.1 | 89.0 | 78.5 | $0.20 | 50 | |
| 17 | gpt-5.6 terra none GPT-5.6 Terra · none | 89.1 | 88.8 | 79.7 | $0.05 | 50 | |
| 18 | gpt-5.6 terra medium GPT-5.6 Terra · medium | 89.1 | 88.9 | 80.1 | $0.05 | 50 | |
| 19 | gpt-5.5 high GPT-5.5 · high | 88.9 | 88.6 | 79.4 | $0.14 | 50 | |
| 20 | gpt-5.6 luna medium GPT-5.6 Luna · medium | 88.7 | 88.6 | 82.0 | $0.02 | 50 | |
| 21 | gpt-5.6 luna high GPT-5.6 Luna · high | 88.7 | 88.6 | 78.9 | $0.03 | 50 | |
| 22 | gpt-5.6 luna low GPT-5.6 Luna · low | 88.6 | 88.5 | 79.4 | $0.02 | 50 | |
| 23 | gpt-5.4 medium GPT-5.4 · medium | 88.4 | 88.3 | 80.1 | $0.06 | 50 | |
| 24 | gpt-5.5 none GPT-5.5 · none | 88.3 | 88.1 | 76.6 | $0.13 | 50 | |
| 25 | gpt-5.6 luna none GPT-5.6 Luna · none | 88.0 | 87.7 | 79.7 | $0.02 | 50 | |
| 26 | gpt-5.5 low GPT-5.5 · low | 87.8 | 87.7 | 78.3 | $0.14 | 50 | |
| 27 | fable 5 high Claude Fable 5 · high | 87.7 | 87.5 | 77.0 | $0.46 | 50 | |
| 28 | gpt-5.4 low GPT-5.4 · low | 87.5 | 87.4 | 78.3 | $0.05 | 50 | |
| 29 | gpt-5.4 none GPT-5.4 · none | 87.5 | 87.4 | 78.2 | $0.05 | 50 | |
| 30 | opus 4.7 max Claude Opus 4.7 · max | 87.2 | 87.3 | 75.0 | $0.21 | 50 | |
| 31 | opus 4.7 high Claude Opus 4.7 · high | 86.6 | 86.8 | 77.8 | $0.17 | 50 | |
| 32 | muse spark 1.1 high Muse Spark 1.1 · high | 86.6 | 86.4 | 77.0 | $0.03 | 50 | |
| 33 | muse spark 1.1 medium Muse Spark 1.1 · medium | 86.4 | 86.2 | 78.4 | $0.02 | 50 | |
| 34 | muse spark 1.1 minimal Muse Spark 1.1 · minimal | 85.7 | 85.7 | 75.8 | $0.02 | 50 | |
| 35 | muse spark 1.1 low Muse Spark 1.1 · low | 85.6 | 85.4 | 75.1 | $0.02 | 50 | |
| 36 | opus 4.8 medium Claude Opus 4.8 · medium | 85.6 | 85.8 | 72.1 | $0.14 | 50 | |
| 37 | opus 4.7 medium Claude Opus 4.7 · medium | 85.5 | 85.6 | 74.0 | $0.33 | 50 | |
| 38 | opus 4.7 xhigh Claude Opus 4.7 · xhigh | 85.5 | 85.6 | 73.1 | $0.17 | 50 | |
| 39 | opus 4.7 low Claude Opus 4.7 · low | 85.5 | 85.6 | 75.6 | $0.13 | 50 | |
| 40 | opus 4.8 max Claude Opus 4.8 · max | 85.4 | 85.4 | 72.9 | $0.18 | 50 | |
| 41 | opus 4.8 xhigh Claude Opus 4.8 · xhigh | 85.3 | 85.2 | 73.0 | $0.16 | 50 | |
| 42 | opus 4.8 high Claude Opus 4.8 · high | 84.9 | 84.9 | 68.8 | $0.15 | 50 | |
| 43 | sonnet 4.6 Claude Sonnet 4.6 · default | 84.5 | 84.6 | 71.9 | $0.10 | 50 | |
| 44 | sonnet 5 Claude Sonnet 5 · default | 84.3 | 84.6 | 71.6 | $0.05 | 50 | |
| 45 | opus 4.8 low Claude Opus 4.8 · low | 83.7 | 84.0 | 67.2 | $0.12 | 50 | |
| 46 | glm 5.2 GLM 5.2 · default | 83.6 | 84.0 | 71.3 | $0.03 | 50 | |
| 47 | deepseek v4 pro DeepSeek V4 Pro · default | 83.1 | 83.5 | 69.1 | $0.0047 | 50 | |
| 48 | gemini 3.1 pro previewTrailing Gemini 3.1 Pro Preview · default | 78.7 | 78.9 | 66.6 | $0.03 | 50 | |
| Filtered benchmark | 87.4 | ||||||
Top model
90.0benchmark
gpt-5.6 sol max
89.7 raw avg · 80.3 p10 floor · $0.37/call
Hardest call
69.4avg
AnthropicExxonMobil
ExxonMobil AI governance and safety review for energy operations with Anthropic
Product demomixedSonnet-generated
Reasoning effort
+1.9pts
GPT-5.6 Luna: none → max
88.0 → 89.9 as effort scales
How this benchmark was made
Methodology
The calls are synthetic, immutable benchmark cases with hidden coaching ground truth.
01
Generate cases
Each case starts with two companies, a call type, duration, quality target, web research, and role notes.
02
Write turn by turn
Half the current calls come from the original GPT-based generator and half from a Claude Sonnet 4.6 generator. Both are written one speaker turn at a time.
03
Judge semantically
Coach models see only the visible case. A judge model sees hidden ground truth, scores eight dimensions, and rolls the sales-critical axes into the benchmark ranking.
- Calls
- 50
- Model configs
- 48
- Judged runs
- 2400
- Origins
- 2
- Axes
- 8
50 sales calls in the benchmark
Browse the calls
Pick a call to see the hidden answer key and model scores.
Discovery95.7
CollibraBerkshire Hathaway
Berkshire Hathaway Data governance discovery across decentralized business units with Collibra
flawed·GPT-generated·33mEasiest
Competitive displacement94.9
StripePave
Pave Pricing and packaging objection call with Stripe
flawed·GPT-generated·18m
Renewal save94.4
TwilioThe Home Depot
The Home Depot Renewal save call after usage and support concerns with Twilio
flawed·GPT-generated·42m
Product demo94.3
MongoDBWayfair
Wayfair Integration deep dive for catalog modernization with MongoDB
excellent·GPT-generated·58m
Discovery94.1
AtlassianDelta Air Lines
Delta Air Lines Enterprise discovery for service management modernization with Atlassian
flawed·GPT-generated·31m
Discovery94.0
WorkdayMcKesson
McKesson HR transformation qualification and stakeholder mapping with Workday
flawed·Sonnet-generated·27m
Discovery93.8
VercelMercury
Mercury First discovery for frontend platform consolidation with Vercel
flawed·GPT-generated·22m
Product demo93.6
Palo Alto NetworksApple
Apple Technical security review for zero trust architecture with Palo Alto Networks
excellent·GPT-generated·66m
QBR93.3
AmplitudeDuolingo
Duolingo Renewal QBR and expansion planning with Amplitude
excellent·GPT-generated·52m
Discovery92.9
WorkdayMcKesson
McKesson HR transformation qualification and stakeholder mapping with Workday
flawed·GPT-generated·27m
Discovery92.4
OpenAICVS Health
CVS Health AI contact-center transformation discovery with OpenAI
excellent·GPT-generated·61m
Discovery91.9
GitHubRippling
Rippling Product-led expansion discovery for developer workflow with GitHub
excellent·GPT-generated·41m
Product demo91.8
AnthropicExxonMobil
ExxonMobil AI governance and safety review for energy operations with Anthropic
mixed·GPT-generated·39m
Discovery91.6
VercelMercury
Mercury First discovery for frontend platform consolidation with Vercel
flawed·Sonnet-generated·22m
Product demo91.4
DatadogLinear
Linear Technical demo for observability and incident response with Datadog
excellent·GPT-generated·34m
Competitive displacement91.4
CloudflareCanva
Canva Competitive displacement discovery for edge security with Cloudflare
flawed·Sonnet-generated·47m
Product demo90.9
CrowdStrikeTarget
Target Security architecture review for endpoint consolidation with CrowdStrike
excellent·GPT-generated·63m
Product demo90.4
ElasticJPMorgan Chase
JPMorgan Chase Technical workshop for search and observability consolidation with Elastic
excellent·GPT-generated·74m
Competitive displacement90.3
StripePave
Pave Pricing and packaging objection call with Stripe
flawed·Sonnet-generated·18m
Discovery89.9
HashiCorpAmazon
Amazon Cloud operating model discussion for internal platform teams with HashiCorp
flawed·GPT-generated·26m
Competitive displacement89.7
ServiceNowFord Motor Company
Ford Motor Company Procurement negotiation for workflow automation with ServiceNow
mixed·GPT-generated·35m
Product demo89.6
MicrosoftCostco Wholesale
Costco Wholesale Proof-of-concept readout for analytics and productivity workflow with Microsoft
mixed·Sonnet-generated·55m
Discovery89.2
NVIDIAWalmart
Walmart Executive discovery for AI infrastructure and store operations with NVIDIA
excellent·GPT-generated·57m
Product demo89.1
MongoDBWayfair
Wayfair Integration deep dive for catalog modernization with MongoDB
excellent·Sonnet-generated·58m
Product demo88.6
CrowdStrikeTarget
Target Security architecture review for endpoint consolidation with CrowdStrike
excellent·Sonnet-generated·63m
Discovery88.6
OpenAICVS Health
CVS Health AI contact-center transformation discovery with OpenAI
excellent·Sonnet-generated·61m
Discovery87.5
GitHubRippling
Rippling Product-led expansion discovery for developer workflow with GitHub
excellent·Sonnet-generated·41m
Product demo87.1
SnowflakeToast
Toast Data platform proof-of-concept kickoff with Snowflake
flawed·GPT-generated·44m
Discovery86.9
AtlassianDelta Air Lines
Delta Air Lines Enterprise discovery for service management modernization with Atlassian
flawed·Sonnet-generated·31m
QBR86.8
OktaSweetgreen
Sweetgreen Executive alignment for identity modernization with Okta
mixed·GPT-generated·38m
Competitive displacement86.4
CloudflareCanva
Canva Competitive displacement discovery for edge security with Cloudflare
flawed·GPT-generated·47m
QBR86.1
OktaSweetgreen
Sweetgreen Executive alignment for identity modernization with Okta
mixed·Sonnet-generated·38m
Product demo86.0
FigmaThe Walt Disney Company
The Walt Disney Company Design collaboration demo with brand and asset workflow discussion with Figma
mixed·GPT-generated·49m
Discovery85.8
NVIDIAWalmart
Walmart Executive discovery for AI infrastructure and store operations with NVIDIA
excellent·Sonnet-generated·57m
Product demo85.6
SnykRunway
Runway Security review before developer-tool rollout with Snyk
mixed·GPT-generated·29m
Discovery84.8
HashiCorpAmazon
Amazon Cloud operating model discussion for internal platform teams with HashiCorp
flawed·Sonnet-generated·26m
Renewal save84.5
SalesforceUnitedHealth Group
UnitedHealth Group Healthcare CRM expansion objection handling with Salesforce
mixed·GPT-generated·46m
Product demo84.3
SnykRunway
Runway Security review before developer-tool rollout with Snyk
mixed·Sonnet-generated·29m
Product demo84.3
FigmaThe Walt Disney Company
The Walt Disney Company Design collaboration demo with brand and asset workflow discussion with Figma
mixed·Sonnet-generated·49m
Renewal save83.7
SalesforceUnitedHealth Group
UnitedHealth Group Healthcare CRM expansion objection handling with Salesforce
mixed·Sonnet-generated·46m
Renewal save82.2
TwilioThe Home Depot
The Home Depot Renewal save call after usage and support concerns with Twilio
flawed·Sonnet-generated·42m
Product demo82.0
DatadogLinear
Linear Technical demo for observability and incident response with Datadog
excellent·Sonnet-generated·34m
Competitive displacement80.8
ServiceNowFord Motor Company
Ford Motor Company Procurement negotiation for workflow automation with ServiceNow
mixed·Sonnet-generated·35m
QBR80.3
AmplitudeDuolingo
Duolingo Renewal QBR and expansion planning with Amplitude
excellent·Sonnet-generated·52m
Product demo79.8
Palo Alto NetworksApple
Apple Technical security review for zero trust architecture with Palo Alto Networks
excellent·Sonnet-generated·66m
Product demo79.6
MicrosoftCostco Wholesale
Costco Wholesale Proof-of-concept readout for analytics and productivity workflow with Microsoft
mixed·GPT-generated·55m
Product demo79.2
SnowflakeToast
Toast Data platform proof-of-concept kickoff with Snowflake
flawed·Sonnet-generated·44m
Discovery73.3
CollibraBerkshire Hathaway
Berkshire Hathaway Data governance discovery across decentralized business units with Collibra
flawed·Sonnet-generated·33m
Product demo71.2
ElasticJPMorgan Chase
JPMorgan Chase Technical workshop for search and observability consolidation with Elastic
excellent·Sonnet-generated·74m
Product demo69.4
AnthropicExxonMobil
ExxonMobil AI governance and safety review for energy operations with Anthropic
mixed·Sonnet-generated·39mHardest
By model family
Does spending more on reasoning help?
For families with multiple reasoning settings, this shows the benchmark score at each level.
Claude Opus 4.7
+1.7 pts
low
85.5
baseline
medium
85.5
+0.0 pts
high
86.6
+1.1 pts
xhigh
85.5
+0.0 pts
max
87.2
+1.7 pts
Claude Opus 4.8
+1.7 pts
low
83.7
baseline
medium
85.6
+1.8 pts
high
84.9
+1.1 pts
xhigh
85.3
+1.6 pts
max
85.4
+1.7 pts
GPT-5.4
+1.8 pts
none
87.5
baseline
low
87.5
+0.0 pts
medium
88.4
+1.0 pts
high
89.2
+1.7 pts
xhigh
89.3
+1.8 pts
GPT-5.5
+0.8 pts
none
88.3
baseline
low
87.8
-0.5 pts
medium
89.1
+0.8 pts
high
88.9
+0.6 pts
xhigh
89.1
+0.8 pts
GPT-5.6 Luna
+1.9 pts
none
88.0
baseline
low
88.6
+0.6 pts
medium
88.7
+0.7 pts
high
88.7
+0.7 pts
xhigh
89.9
+1.9 pts
max
89.9
+1.9 pts
GPT-5.6 Sol
+0.5 pts
none
89.6
baseline
low
89.8
+0.2 pts
medium
89.3
-0.3 pts
high
89.4
-0.1 pts
xhigh
89.7
+0.2 pts
max
90.0
+0.5 pts
GPT-5.6 Terra
+0.9 pts
none
89.1
baseline
low
89.9
+0.8 pts
medium
89.1
-0.0 pts
high
89.7
+0.6 pts
xhigh
89.7
+0.6 pts
max
90.0
+0.9 pts
Muse Spark 1.1
+0.9 pts
minimal
85.7
baseline
low
85.6
-0.1 pts
medium
86.4
+0.7 pts
high
86.6
+0.9 pts