Head-to-head comparison

GPT-6 Astra vs. Mistral Medium 3.5

Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.

shared result series: 159
distinct benchmark metrics: 15
latest retrieval: 2026-10-15

First modelSecond model

Specifications

General specifications

Release date, availability, context window, architecture, and published parameter counts at a glance. Undisclosed details remain marked as unavailable.

Specifications for GPT-6 Astra and Mistral Medium 3.5
Attribute	GPT-6 Astra	Mistral Medium 3.5
Provider	OpenAI	Mistral AI
Released	Sep 3, 2026	Apr 28, 2026
Availability	Limited access	Active
Model type	Proprietary	Open Weights
Parameters	Not published	128B
Architecture	Not published	Dense
Context window	1,050,000 tokens	262,144 tokens
Knowledge cutoff	April 30, 2026	Not published
Technical sources	Model, Context	Model, Context

Pricing

API pricing compared

The table compares official standard API rates per 1M tokens, including published cache pricing. Batch, fast, priority, regional surcharges, tool calls, and cloud platform pricing are excluded.

API pricing for GPT-6 Astra and Mistral Medium 3.5
Price type	GPT-6 Astra	Mistral Medium 3.5
API input	$10 ≤272K / $20 >272K	$1.5
API output	$50 ≤272K / $75 >272K	$7.5
Cache read	$1 ≤272K / $2 >272K	Not listed
Cache write (5 min.)	$12.5 ≤272K / $25 >272K	Not listed
Cache write (1 hr.)	Not listed	Not listed
Price verified	09/22/2026 Source	09/22/2026 Source

GPT-6 AstraMistral Medium 3.5

API inputUSD per 1M tokens, base tier

GPT-6 Astra$10

Mistral Medium 3.5$1.5

API outputUSD per 1M tokens, base tier

GPT-6 Astra$50

Mistral Medium 3.5$7.5

Key to the marks:better valuebetter, but not conclusivetie

Pricing by provider

Direct developer pricing and provider endpoint tariffs routed through OpenRouter are listed separately. Prices are in USD per 1M tokens. Context-dependent direct tiers plus regional, Flex, Priority, and other OpenRouter tariffs remain individually identifiable.

Provider pricing for GPT-6 Astra and Mistral Medium 3.5
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
GPT-6 Astra	OpenAI	Direct from developerUp to 272,000 context tokens	$10	$50	Source09/22/2026
	OpenAI	Direct from developerAbove 272,000 context tokens	$20	$75	Source09/22/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-west-2	$11	$55	Source10/19/2026
	Azure	Via OpenRouterazure	$10	$50	Source10/19/2026
	Azure	Via OpenRouterazure/us	$11	$55	Source10/19/2026
	OpenAI	Via OpenRouteropenai	$10	$50	Source10/19/2026
	OpenAI	Via OpenRouteropenai/fast	$20	$100	Source10/19/2026
	OpenAI	Via OpenRouteropenai/flex	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	OpenAI	Via OpenRouteropenai/ultrafast	$60	$300	Source10/19/2026
Mistral Medium 3.5	Mistral AI	Direct from developerStandard API	Lowest listed price: $1.5	Lowest listed price: $7.5	Source09/22/2026
	Mistral	Via OpenRoutermistral	Lowest listed price: $1.5	Lowest listed price: $7.5	Source10/19/2026
	Mistral	Via OpenRoutermistral/eu	$1.65	$8.25	Source10/19/2026
	Mistral	Via OpenRoutermistral/zdr	Lowest listed price: $1.5	Lowest listed price: $7.5	Source10/19/2026

Performance

Shared benchmarks

Only published results with the same benchmark version, task, metric, and comparison cohort are paired. Different reasoning levels, agents, harnesses, or output limits appear directly in the table.

Capability profile from matched benchmarks

Each axis averages directly matched benchmark families. A value of 100 means the stronger value within this pair, not a universal quality score.

Not enough directly matched categories for a reliable radar

Only 0 of at least 5 categories have comparable, direct model measurements. The benchmark tables below show the available evidence without filling missing axes.

GPT-6 AstraMistral Medium 3.5

Code MigrationOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra67.74 %

Mistral Medium 3.55.127 %

EMBOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra71.7 %

Mistral Medium 3.511.185 %

Finance Agent (v2)Overall. Higher is better. Test setup not documented as identical

GPT-6 Astra53.54 %

Mistral Medium 3.532.103 %

Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. Test setup not documented as identical

GPT-6 Astra5.417 %

Mistral Medium 3.50.417 %

Legal Research BenchOverall · All-pass. Higher is better. Test setup not documented as identical

GPT-6 Astra39.423 %

Mistral Medium 3.59.135 %

LMArena Code Arena WebDevOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra1,780.244

Mistral Medium 3.51,263.383

LMArena Text, Style ControlOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra1,476.716

Mistral Medium 3.51,426.605

LMArena Vision, Style ControlOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra1,285.988

Mistral Medium 3.51,200.386

MedCodeOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra48.486 %

Mistral Medium 3.533.752 %

MedScribeOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra87.908 %

Mistral Medium 3.567.728 %

ProofBenchOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra99 %

Mistral Medium 3.59 %

SAGEOverall. Higher is better. Test setup not documented as identical

GPT-6 Astra46.367 %

Mistral Medium 3.537.613 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for GPT-6 Astra and Mistral Medium 3.5
Benchmark and task	GPT-6 Astra	Mistral Medium 3.5	Source
Code MigrationOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value67.74 %	5.127 %	Vals AI Source details Data as of: 10/12/2026
EMBOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value71.7 %	11.185 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value53.54 %	32.103 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkOverall · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value5.417 %	0.417 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchOverall · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value39.423 %	9.135 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevOverall Test details Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 5,746 published pairwise votes.Mistral Medium 3.5: Arena rating from 2,320 published pairwise votes.	Higher measured value1,780.244 95% confidence interval: ±10.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,263.383 95% confidence interval: ±14.784The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Text, Style ControlOverall Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 11,674 published pairwise votes.Mistral Medium 3.5: Arena rating from 11,793 published pairwise votes.	Higher measured value1,476.716 95% confidence interval: ±6.385The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,426.605 95% confidence interval: ±6.399The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlOverall Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Vision, style controlGPT-6 Astra: Arena rating from 3,590 published pairwise votes.Mistral Medium 3.5: Arena rating from 5,375 published pairwise votes.	Higher measured value1,285.988 95% confidence interval: ±10.782The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,200.386 95% confidence interval: ±9.604The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
MedCodeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	Higher measured value48.486 %	33.752 %	Vals AI Source details Data as of: 10/13/2026
MedScribeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	Higher measured value87.908 %	67.728 %	Vals AI Source details Data as of: 10/13/2026
ProofBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value99 %	9 %	Vals AI Source details Data as of: 10/12/2026
SAGEOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	Higher measured value46.367 %	37.613 %	Vals AI Source details Data as of: 10/08/2026
Tax Agent BenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value20.717 %	4.482 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value87.266 %	38.951 %	Vals AI Source details Data as of: 09/28/2026
Vibe Code Bench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Harness: OpenHands, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value89.594 %	2.888 %	Vals AI Source details Data as of: 10/12/2026
Code MigrationCLI Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value67.463 %	4.991 %	Vals AI Source details Data as of: 10/12/2026
EMBComps Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value77.953 %	18.088 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Adjustments Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value48.318 %	27.312 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value88.768 %	64.493 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchAdministrative / Regulatory · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value53.03 %	7.576 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevWebdev Test details Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 5,746 published pairwise votes.Mistral Medium 3.5: Arena rating from 2,320 published pairwise votes.	Higher measured value1,780.244 95% confidence interval: ±10.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,263.383 95% confidence interval: ±14.784The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Text, Style ControlBusiness, management and finance Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 2,061 published pairwise votes.Mistral Medium 3.5: Arena rating from 2,257 published pairwise votes.	Higher measured value1,472.34 95% confidence interval: ±13.336The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,430.463 95% confidence interval: ±12.817The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlChinese Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Vision, style controlGPT-6 Astra: Arena rating from 230 published pairwise votes.Mistral Medium 3.5: Arena rating from 288 published pairwise votes.	Higher measured value1,357.126 95% confidence interval: ±39.572The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,216.486 95% confidence interval: ±38.816The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
SAGEAutomata Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	Higher measured value37.302 %	30.159 %	Vals AI Source details Data as of: 10/08/2026
Tax Agent BenchControversy & Precedence Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value13.333 %	3.333 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Easy Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value100 %	66.667 %	Vals AI Source details Data as of: 09/28/2026
Code MigrationCOBOL Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value68.571 %	5.536 %	Vals AI Source details Data as of: 10/12/2026
EMBDataroom Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value79.319 %	13.403 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)All-Pass Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value40.252 %	21.632 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	No direct winner0 %	No direct winner0 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: weighted partial-credit scoreScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value87.482 %	67.472 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevWebdev-html Test details Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 1,129 published pairwise votes.Mistral Medium 3.5: Arena rating from 259 published pairwise votes.	Higher measured value1,780.13 95% confidence interval: ±22.887The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,307.357 95% confidence interval: ±41.479The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Text, Style ControlChinese Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 877 published pairwise votes.Mistral Medium 3.5: Arena rating from 620 published pairwise votes.	Higher measured value1,514.537 95% confidence interval: ±19.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,446.79 95% confidence interval: ±23.875The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlCreative writing vision Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Vision, style controlGPT-6 Astra: Arena rating from 367 published pairwise votes.Mistral Medium 3.5: Arena rating from 342 published pairwise votes.	Higher measured value1,289.081 95% confidence interval: ±32.749The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,196.259 95% confidence interval: ±33.621The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
SAGECalculus Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	Higher measured value79.938 %	2.778 %	Vals AI Source details Data as of: 10/08/2026
Tax Agent BenchControversy & Precedence Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value58.616 %	18.379 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Hard Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value87.778 %	15.556 %	Vals AI Source details Data as of: 09/28/2026
Code MigrationCode Quality Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: mean reviewer code-quality ratingScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value4.469 / 5	2.987 / 5	Vals AI Source details Data as of: 10/12/2026
EMBDCF Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value64.483 %	2.218 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Comparables Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value37.758 %	31.512 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkArbitration International Dispute Resolution · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value86.022 %	70.776 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchBusiness & Commercial · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value32.558 %	9.302 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevWebdev-react Test details Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDev Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Code Arena WebDevGPT-6 Astra: Arena rating from 3,422 published pairwise votes.Mistral Medium 3.5: Arena rating from 1,977 published pairwise votes.	Higher measured value1,804.405 95% confidence interval: ±13.548The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,253.399 95% confidence interval: ±16.155The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Text, Style ControlCoding Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 2,659 published pairwise votes.Mistral Medium 3.5: Arena rating from 3,298 published pairwise votes.	Higher measured value1,543.161 95% confidence interval: ±11.756The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,477.902 95% confidence interval: ±10.821The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlDiagram Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Vision, style controlGPT-6 Astra: Arena rating from 861 published pairwise votes.Mistral Medium 3.5: Arena rating from 1,540 published pairwise votes.	Higher measured value1,310.509 95% confidence interval: ±20.739The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,242.126 95% confidence interval: ±16.169The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
SAGELinear Algebra Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	Higher measured value62.117 %	41.411 %	Vals AI Source details Data as of: 10/08/2026
Tax Agent BenchCurrent & Temporal Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value29.944 %	6.5 %	Vals AI Source details Data as of: 10/12/2026
Terminal-Bench 2.1Medium Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value86.061 %	49.697 %	Vals AI Source details Data as of: 09/28/2026
EMBLBO Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value61.825 %	9.61 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Disclosure Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value61.684 %	40.223 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkArbitration International Dispute Resolution · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	No direct winner0 %	No direct winner0 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchBusiness & Commercial · Weighted partial credit Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: weighted partial-credit scoreScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value83.821 %	71.304 %	Vals AI Source details Data as of: 10/12/2026
LMArena Text, Style ControlCreative writing Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 2,620 published pairwise votes.Mistral Medium 3.5: Arena rating from 2,081 published pairwise votes.	Higher measured value1,449.578 95% confidence interval: ±12.436The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,400.583 95% confidence interval: ±13.731The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlEnglish Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Vision, style controlGPT-6 Astra: Arena rating from 1,205 published pairwise votes.Mistral Medium 3.5: Arena rating from 2,316 published pairwise votes.	Higher measured value1,288.398 95% confidence interval: ±17.904The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,216.805 95% confidence interval: ±13.853The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
SAGEPartial Differential Equations Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 1, Top P: 0.95, Output limit: 30,000 tokens	39.683 %	Higher measured value41.27 %	Vals AI Source details Data as of: 10/08/2026
Tax Agent BenchCurrent & Temporal Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value66.224 %	29.444 %	Vals AI Source details Data as of: 10/12/2026
EMBM&A Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value83.066 %	10.694 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Earnings Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value69.474 %	42.125 %	Vals AI Source details Data as of: 10/12/2026
Harvey's Legal Agent BenchmarkBanking Finance · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value88.104 %	78.439 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchCivil Litigation · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value40.741 %	11.111 %	Vals AI Source details Data as of: 10/12/2026
LMArena Text, Style ControlEnglish Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Text, style controlGPT-6 Astra: Arena rating from 4,703 published pairwise votes.Mistral Medium 3.5: Arena rating from 5,655 published pairwise votes.	Higher measured value1,485.113 95% confidence interval: ±9.08The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,443.972 95% confidence interval: ±8.479The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlHomework Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterGPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style control Mistral Medium 3.5: Snapshot: mistral-medium-3.5, Harness: Arena Vision, style controlGPT-6 Astra: Arena rating from 480 published pairwise votes.Mistral Medium 3.5: Arena rating from 805 published pairwise votes.	Higher measured value1,346.15 95% confidence interval: ±26.584The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,217.674 95% confidence interval: ±22.051The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
Tax Agent BenchFact-Pattern Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value27.907 %	2.326 %	Vals AI Source details Data as of: 10/12/2026
EMBMisc Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterGPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens Mistral Medium 3.5: Reasoning: high, Temperature: 0.7, Top P: 0.95, Output limit: 80,000 tokens	Higher measured value59.814 %	14.103 %	Vals AI Source details Data as of: 10/12/2026
Key to the marks:better valuebetter, but not conclusivetie

Our model tests

Our tests

SVG short test

One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.

Shared test round: 4ae5d8eeb9a9

GPT-6 Astra

Unavailable in this round

Mistral Medium 3.5

Earlier tests

Pelican on a bicycle

A community classic for free-form SVG drawing.

GPT-6 Astra

Reasoning (requested): high

Current attempt

Run date: Sep 7, 2026
API provider: openrouter
Requested API model: openai/gpt-6-astra
API model reported in response: openai/gpt-6-astra
Reasoning setting: high
Pinned endpoint: OpenAI | openai/gpt-6-astra-20260903
Temperature: Provider default
Token limit including reasoning: 8,192 tokens
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating
Response time: 141.7 s

Visually reviewed. This individual test does not establish overall model quality.

Mistral Medium 3.5

Reasoning (requested): high

Current attempt

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: mistralai/mistral-medium-3-5
API model reported in response: mistralai/mistral-medium-3-5
Reasoning setting: high
Pinned endpoint: Mistral | mistralai/mistral-medium-3.5-20260430
Temperature: Provider default
Token limit including reasoning: 8,192 tokens
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating
Response time: 38.7 s

Visually reviewed. This individual test does not establish overall model quality.

Earlier attempts (1)

Sep 7, 2026API request failedReasoning setting: high

Prompt and test conditions

Generate an SVG of a pelican riding a bicycle

Token limit including reasoning: 8,192 tokens

Task origin (Simon Willison)

How to read this comparison

Winning one benchmark is not an overall verdict

Your workload, budget, and required context length matter most. A coding benchmark says little about visual understanding or agent performance.

The standard price table uses direct developer rates. The provider comparison labels OpenRouter endpoints, regions, and special tariffs separately.

Every result links to its measurement source and retrieval date. Labels distinguish vendor reports, official benchmark leaderboards, and independent evaluations. Disputed or archived results are excluded.

More matchups

Compare GPT-6 Astra and Mistral Medium 3.5 with other leading models using the same data and benchmark logic.

View the complete LLM comparison

GPT-6 Astra vs. Mistral Medium 3.5

Change model selection

General specifications

API pricing compared

Shared benchmarks

Capability profile from matched benchmarks

Our tests

SVG short test

GPT-6 Astra

Mistral Medium 3.5

Pelican on a bicycle

Winning one benchmark is not an overall verdict

Related LLM comparisons