Gemini 4 Argon vs GPT-6.1 Sol
They share the exact same $2/$10 list price through Nov 30 — why do real task bills and long-context costs diverge?
The short answer
| At a glance | Gemini 4 Argon | GPT-6.1 Sol |
|---|---|---|
| List priceinput / output per 1M tokens | $2 / $10 | $2 / $10 |
| Benchmark task costArtificial Analysis, per Intelligence-Index task | $1.99 | $0.72 (better) |
| Intelligence IndexAA v4.3.2, higher is better | 53 (better) | 52 |
On paper, Gemini 4 Argon and GPT-6.1 Sol have identical standard, batch, and flex list prices through November 30, 2026 ($2.00 in / $10.00 out / $0.10 cached read per MTok; $1.00 / $5.00 on Batch and Flex). In practice, three pricing mechanics separate them: (1) Reasoning verbosity — on Artificial Analysis's Intelligence Index v4.3.2, Argon edges Sol on quality (Index 53 vs 52, AutomationBench-AA 78% vs 65%, SciCode 62% vs 54%) but emits ~1.6x–4.7x more output tokens, making Argon's measured benchmark task cost $1.99 vs Sol's $0.72; (2) Long-context cliff — above 272K input tokens per request, GPT-6.1 Sol doubles input to $4/$15 while Argon stays flat at $2/$10 (until Dec 1); and (3) The Dec 1 sunset — starting December 1, 2026, Argon's introductory price expires and doubles to $4/$20, making GPT-6.1 Sol half the price on short/medium prompts.
Choose Gemini 4 Argon
Agentic SaaS automation (AutomationBench-AA 78% vs 65%), scientific/coding reasoning (SciCode 62% vs 54%, HLE 57% vs 53%), prompts over 272K tokens (no long-context surcharge through Nov 30), or long-decode tasks over 128K output tokens.
Choose GPT-6.1 Sol
Task-cost efficiency ($0.72 vs $1.99 per AA task due to lower reasoning token verbosity), PDF document Q&A (GDP.pdf 31% vs 22%), predictable $2/$10 pricing after December 1, or when explicit cache writes and Fast/Ultrafast serving tiers are needed.
Either — decide by evidence you generate
Batch/Flex backfills before Nov 30: both offer $1.00/$5.00 Batch and Flex tiers — test whether Argon's medium/low thinking level matches Sol's token count on your prompts.
Key facts
- Identical headline rates through 2026-11-30: $2.00 input / $10.00 output / $0.10 cached read per MTok on Standard, and $1.00 / $5.00 / $0.05 on Batch and Flex.
- Two opposite price cliffs: GPT-6.1 Sol steps up to $4.00 / $15.00 on any request above 272,000 input tokens; Gemini 4 Argon has NO prompt-size cliff (flat to 1M), but its introductory price expires on 2026-12-01 and steps up to $4.00 / $20.00 across all prompt sizes.
- Per-task cost divergence at identical token prices (AA v4.3.2): Argon costs $1.99 per Intelligence-Index task vs GPT-6.1 Sol's $0.72 (Max) / $0.32 (High), because Argon spends more tokens thinking (62k output tokens/task vs 38k on Sol Max and 13k on Sol High).
- Independent same-harness quality (AA v4.3.2): Argon leads overall Intelligence Index (53 vs 52 Sol Max / 50 Sol High), AutomationBench-AA (78% vs 65%), SciCode (62% vs 54%), HLE (57% vs 53%), GDPval-AA (1624 vs 1592), and Terminal-Bench 4.0 (57% vs 56%); Sol leads AA-Briefcase (1557 vs 1487) and GDP.pdf (31% vs 22%).
- Caching trade-off: Argon's implicit caching charges $0 for cache writes (95% discount on hits above 4,096 tokens, no explicit TTL control); GPT-6.1 Sol supports explicit prompt caching and bills cache writes at $2.50/MTok ($5.00 above 272K).
Official pricing, every tier
USD per 1M tokens from vendor pricing pages. 'Not published' means exactly that — we don't extrapolate.
| Service tier | Gemini 4 Argon | GPT-6.1 Sol |
|---|---|---|
| Standarddefault on-demand serving | in $2 · cached $0.10 · out $10 implicit cache only (no write fee) | in $2 · cached $0.10 · out $10 cache write $2.5 · >272K in/req |
| Batchasync jobs, 50% off | in $1 · cached $0.05 · out $5 implicit cache only (no write fee) | in $1 · cached $0.05 · out $5 cache write $1.25 · >272K in/req |
| Flexslower/cheaper serving, 50% off | in $1 · cached $0.05 · out $5 implicit cache only (no write fee) | in $1 · cached $0.05 · out $5 cache write $1.25 · >272K in/req |
| Fastpriority speed, 2x (ex-Priority) | Not published | in $4 · cached $0.20 · out $20 cache write $5 · >272K in/req |
| Ultrafast6x standard price | Not published | in $12 · cached $0.60 · out $60 cache write $15 · >272K in/req |
| Long-context band | Gemini 4 Argon | GPT-6.1 Sol |
|---|---|---|
| Long-context ratesapplies per request above the model threshold | Flat pricing same rates at any input size (up to context window) | in $4 · cached $0.20 · out $15 cache write $5 · >272K in/req |
USD per 1M tokens. Every input token is billed in exactly one bucket — uncached, cached read, or cache write; a write replaces the input rate for those tokens (official rule), so write rates are per tier as listed.OpenAI long context = input > 272,000 tokens per single request. · OpenAI pricing (verified 2026-10-09) · Google Gemini 4 Argon shows introductory rates valid through 2026-11-30 ($4/$20 standard starting 2026-12-01) and uses implicit caching only (95% discount on cache hits, no explicit write fee). · Google Gemini pricing (verified 2026-10-13)
Tier ≠ quality change
Run this comparison on your own numbers
Preloaded with a workload typical for this comparison. Change anything — results recompute locally.
Token-based workload estimate only; it does not measure quality or cost per successful task.
| Model | Per request | Monthly |
|---|---|---|
| Gemini 4 Argon cheapest | $0.45 | $4,500.00 |
| GPT-6.1 Sol long ctx | $0.88 | $8,800.00 |
Monthly spread between the cheapest and priciest selected model: $4,300.00.
Breakdown: Gemini 4 Argon · standard tier · standard context
- Fresh input (uncached): 200,000 tok × $2/M$0.40
- Cached input (reads): 100,000 tok × $0.1/M$0.01
- Cache writes: 0 tok × $0/M$0.00
- Output (incl. reasoning tokens): 4,000 tok × $10/M$0.04
- Per request$0.45
Rates: official pricing · verified 2026-10-13
- Per-token arithmetic only: it does not model quality differences or cross-vendor output-token-count differences for the same task.
- Reasoning tokens are billed as output tokens at the chosen model's output rate.
- Long-context band is decided per request by total input tokens vs the provider threshold, not by monthly volume.
- Cache accounting uses the official mutually-exclusive buckets: each input token is billed once, as uncached input, cached read, OR cache write — writes replace the input rate for those tokens, never add to it.
- Claude rows bill the 5-minute-TTL cache write; the published 1-hour TTL ($8/MTok standard) is not modeled (no TTL input).
- Rows flagged 'derived' are computed from a stated official rule (e.g. Anthropic batch -50%), not read verbatim from a price table.
- Missing official prices render as Unsupported — we never apply a guessed multiplier (e.g., GPT-6 Sol has no Ultrafast tier).
What the evidence actually says
Independent and vendor-run evals, each with publisher, runner, and harness. Numbers from different runs are never merged.
| Benchmark | Score | Effort | Harness | Evidence |
|---|---|---|---|---|
| DeepSWEv1.1 | ||||
| gpt-6.1-sol | 75.2 % | high | OpenAI announcement evals | Vendor-reported |
| AutomationBench1.0.6 | ||||
| gpt-6.1-sol | +2.2 pp vs Claude Opus 5.5; +4.8 pp vs GPT-6 Sol | medium | OpenAI announcement (underlying runner not stated — AutomationBench is run/reported by Zapier per Anthropic's footnote) | Vendor-reported |
| OSWorld2.0 offline set, v2026.08.08 | ||||
| gpt-6.1-sol | within 2.1 pp of GPT-6 Astra; +7 pp vs GPT-6 Sol | max | OpenAI announcement evals | Vendor-reported |
| Terminal-Bench-Science cost per task0.1 | ||||
| gpt-6.1-sol | 5.47 usd-per-task | unspecified | OpenAI announcement evals | Vendor-reported |
| Factuality error rateunspecified | ||||
| gpt-6.1-sol | 7.7 error-rate-% | low | OpenAI announcement evals | Vendor-reported |
| GDP.pdf (professional PDF Q&A)unspecified | ||||
| gpt-6.1-sol | scores higher than Opus 5.5 with fallbacks at less than half the cost per task; near Astra at ~1/5 cost per task | multiple settings | OpenAI announcement evals | Vendor-reported |
| AA Intelligence Indexv4.3.2 | ||||
| gpt-6.1-sol | 52 index-points | config 'Sol (Max)' per AA | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 53 index-points | config 'Gemini 4 Argon (High)' per AA | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| Terminal-Bench4.0 | ||||
| gpt-6.1-sol | 56 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 57 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| SciCodeunspecified | ||||
| gpt-6.1-sol | 54 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 62 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| GDPval-AAv2.1 | ||||
| gpt-6.1-sol | 1592 elo | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 1624 elo | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| AA-Briefcasev1.1 | ||||
| gpt-6.1-sol | 1557 elo | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 1487 elo | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| Humanity's Last Examunspecified | ||||
| gpt-6.1-sol | 53 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 57 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| AutomationBench-AAunspecified | ||||
| gpt-6.1-sol | 65 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 78 % | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| AA cost per Intelligence-Index taskv4.3.2 | ||||
| gpt-6.1-sol | 0.72 usd-per-task | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| gemini-4-argon | 1.99 usd-per-task | — | Artificial Analysis Intelligence Index v4.3.2 (independent, same harness for all models) | Third-party |
| Vals Index (DeepMind table)Oct 2026 | ||||
| gemini-4-argon | 68.9 % | — | Google DeepMind Model Card evaluation table (finance, legal, coding, and tax domains) | Vendor-reported |
| GraphWalks (256K–1M long context)BFS F1 | ||||
| gemini-4-argon | 84.2 % | — | Google DeepMind Model Card evaluation table | Vendor-reported |
| LABBench 2v2 | ||||
| gemini-4-argon | 88.8 % | — | Google DeepMind Model Card evaluation table | Vendor-reported |
Each row records who published and who ran it (see the Harness column); numbers compare only within a documented run.
Sources: OpenAI — Introducing GPT-6.1 Sol (DevDay 2026) · Artificial Analysis (independent) — Independent same-harness comparison: GPT-6.1 Sol vs Claude Opus 5.5 (Intelligence Index v4.3.2) · Google DeepMind — Gemini 4 Argon Model Card (October 2026 evaluation table)
Which model for which workload
Confidence labels: Measured = same-harness evidence · Inferred = reasoned from documented evidence · Insufficient evidence = no comparable public data.
| Workload | Our pick | Confidence | Why |
|---|---|---|---|
| Daily coding tasks | Gemini 4 Argon | Measured | Leads same-harness AA SciCode (62% vs 54%) and Terminal-Bench 4.0 (57% vs 56%) at the same $2/$10 list price through Nov 30, though Sol uses fewer reasoning tokens per task. |
| Complex repository maintenance | Gemini 4 Argon | Measured | Stronger SciCode (62% vs 54%), 77.9% DeepSWE v1.1 in DeepMind's Model Card, and no 272K input surcharge when feeding whole-repo context. |
| Agent workflows | Gemini 4 Argon | Measured | Argon leads AA AutomationBench-AA by 13 points (78% vs 65%) and GDPval-AA (1624 vs 1592) in the same harness. |
| Long-context tasks | Gemini 4 Argon | Price-derived | Above 272K input tokens per request, Sol doubles to $4/$15 while Argon stays flat at $2/$10 through Nov 30 (and even after Dec 1 ties on $4 input with zero cache-write fees). |
| Quality-first | Gemini 4 Argon | Measured | Ahead on AA Intelligence Index (53 vs 52), AutomationBench-AA (78% vs 65%), SciCode (62% vs 54%), and HLE (57% vs 53%). |
| Budget-sensitive / high volume | GPT-6.1 Sol | Measured | $0.72 vs $1.99 per AA task despite identical $2/$10 token rates, and Sol keeps its $2/$10 list price after Argon's Dec 1 step-up to $4/$20. |
| Offline batch jobs | GPT-6.1 Sol | Measured | Both price Batch and Flex at $1/$5 through Nov 30, but Sol's lower output verbosity and permanent $1/$5 batch rate after Dec 1 keep bulk bills lower. |
| Latency-sensitive production | GPT-6.1 Sol | Price-derived | Sol offers published Fast (2x) and Ultrafast (6x) tiers and measured AA TTFT; Argon's Priority tier (1.8x) has no published AA speed benchmark yet. |
Frequently asked questions
Are Gemini 4 Argon and GPT-6.1 Sol the same price?
Through November 30, 2026, yes on headline list rates: both cost $2.00 per 1M input tokens, $10.00 per 1M output tokens, and $0.10 per 1M cached-input tokens on the Standard tier ($1.00 / $5.00 on Batch and Flex). Starting December 1, 2026, Gemini 4 Argon steps up to $4.00 / $20.00 ($0.20 cached), making it 2x GPT-6.1 Sol's short-context price.
Why did Gemini 4 Argon cost $1.99 per task vs GPT-6.1 Sol's $0.72 at the exact same token price?
Reasoning token verbosity. On Artificial Analysis's Intelligence Index v4.3.2, Gemini 4 Argon (High) emitted an average of 62k output tokens per task (including 36k reasoning tokens), whereas GPT-6.1 Sol emitted 38k on Max ($0.72/task) and 13k on High ($0.32/task). Because thinking tokens are billed at the $10/MTok output rate, emitting more reasoning tokens directly multiplies per-task cost. Lowering Argon's thinking_level to medium or low can reduce token use.
What happens on requests over 272,000 input tokens?
GPT-6.1 Sol switches to its long-context band above 272K input tokens per request, doubling input to $4.00/MTok and raising output to $15.00/MTok. Gemini 4 Argon has flat pricing across its entire 1,048,576-token unified window ($2.00/$10.00 through Nov 30; $4.00/$20.00 from Dec 1). On a 350K-input request before Nov 30, Argon is half the input price and 33% cheaper on output.
Which model is stronger on coding and agent benchmarks?
In Artificial Analysis's independent same-harness evaluation (v4.3.2), Gemini 4 Argon leads GPT-6.1 Sol on the overall Intelligence Index (53 vs 52), AutomationBench-AA (78% vs 65%), SciCode (62% vs 54%), Humanity's Last Exam (57% vs 53%), and GDPval-AA (1624 vs 1592). GPT-6.1 Sol leads on AA-Briefcase (1557 vs 1487) and professional PDF reasoning (GDP.pdf 31% vs 22%).
Why does the calculator show 'Unsupported' on Gemini 4 Argon when cache writes are above 0?
Google's Gemini 4 Argon API supports implicit context caching only — explicit cache creation (cachedContents) is not supported, and Google does not bill a separate cache-write fee. Set 'Cache writes / request' to 0 when comparing Gemini 4 Argon; any cached input tokens you enter are billed at the 95%-discounted $0.10/MTok read rate.
Related
Sources & freshness
- Google — Gemini 4 Argon API model documentation (gemini-4-argon specs, 1M unified context, long decode)https://ai.google.dev/gemini-api/docs/models/gemini-4-argon · published 2026-10-13 · accessed 2026-10-13Model code gemini-4-argon: unified 1,048,576 input+output token window, max output 65,536 synchronous default or up to 1,048,576 with background long decode. Thinking levels low/medium/high (default medium). Implicit caching only (95% discount above 4,096 tokens).
- OpenAI — OpenAI API model catalog (model IDs, context windows, reasoning efforts, capabilities)https://developers.openai.com/api/docs/models · accessed 2026-10-09GPT-6 Sol is NOT listed in the catalog front page at access time (still priced on the pricing page). Knowledge cutoffs: Apr 30 2026 (Astra/6.1 Sol), May 18 2026 (Luna).
- OpenAI — OpenAI API pricing (GPT-6 family tier matrix: standard/batch/flex/fast/ultrafast, short/long context, cache)https://developers.openai.com/api/docs/pricing · accessed 2026-10-09Long context defined as >272K input tokens per request (2x input and cache rates, 1.5x output for the whole request). 'Priority processing' renamed 'Fast mode' on 2026-07-30. Flex and Batch are half of Standard for the GPT-6 family, Fast is 2x, Ultrafast is 6x. Cache writes cost 1.25x the uncached input rate, reads 0.1x (0.05x on GPT-6.1 Sol). Cache-write rates are captured on every published tier row, not Standard only.
- Google — Gemini Developer API pricing (Gemini 4 Argon standard/batch/flex/priority, context caching)https://ai.google.dev/gemini-api/docs/pricing · accessed 2026-10-13Gemini 4 Argon (gemini-4-argon) paid tier introductory pricing through 2026-11-30: Standard $2.00 input / $10.00 output / $0.10 context caching per 1M tokens ($4.00 / $20.00 / $0.20 starting 2026-12-01). Batch and Flex are 50% off ($1.00 / $5.00 / $0.05 through 2026-11-30; $2.00 / $10.00 / $0.10 starting 2026-12-01). Priority tier is 1.8x ($3.60 / $18.00 / $0.18 through 2026-11-30). Implicit caching only (explicit caching is not supported; no cache-write fee). Not available on the Free tier.
- OpenAI — Introducing GPT-6.1 Sol (DevDay 2026)https://openai.com/index/introducing-gpt-6-1-sol · published 2026-09-29 · accessed 2026-10-09Self-reported evals (OpenAI harness). Includes per-task cost quotes (Terminal-Bench-Science). Announces GPT-6.1 Sol Ultrafast 'coming soon' in Codex. Availability: ChatGPT Work + Codex, not yet in Chat.
- Artificial Analysis (independent) — Independent same-harness comparison: GPT-6.1 Sol vs Claude Opus 5.5 (Intelligence Index v4.3.2)https://artificialanalysis.ai/models/comparisons/gpt-6-1-sol-vs-claude-opus-5-5 · accessed 2026-10-10Independent same-harness evaluation of both models. Cost-per-task = weighted average per Intelligence Index task; reflects measured token verbosity. Effort configs as labeled by AA ('Sol (Max)' vs 'Opus 5.5 (Max, Default Fallback)').
- Google DeepMind — Gemini 4 Argon Model Card (October 2026 evaluation table)https://deepmind.google/models/model-cards/gemini-4-argon/ · published 2026-10-13 · accessed 2026-10-13Official evaluation table comparing Gemini 4 Argon, GPT-6 Astra, Claude Fable 5.1, and Claude Opus 5.5 across Vals Index (68.9% vs 67.0% Opus), DeepSWE v1.1 (77.9% vs 74.2% Opus), Terminal-Bench 4.0 (57.4% vs 66.4% Opus), GraphWalks 256k-1M (84.2% vs 66.8% Opus), LABBench 2 (88.8% vs 73.1% Opus), and RiemannBench (76.0% vs 69.6% Opus). Knowledge cutoff March 2026.
Pricing data last verified 2026-10-13. OpenAI ships roughly weekly — re-verify before making spend commitments. See the methodology page for update cadence and stale-data flags.