Head to head · LLM inference · October 2026 research run

Gemini 4 Argon vs StepFun API

Gemini 4 Argon scores 65.3 (B) on agent readiness against StepFun API's 45.6 (E), and leads in 6 of 7 scored categories. Both do llm inference.

Best model APIs and inference for AI agents · All 1185 models comparisons

Which one, for what

Gemini 4 Argon B

Best for Agents doing advanced coding and knowledge work on the paid Gemini API that want Gemini 4 reasoning with fewer thinking tokens per turn than Carbon.

Ahead on

  • Reliability, 45 against 28
  • Schema & documentation, 85 against 53
  • Agent ergonomics, 93 against 68
  • Security & auth, 55 against 41
  • Maintenance & community, 87 against 50
  • Transparency & trust, 73 against 57

Watch for

Paid tier only, so an agent can't start on a free key

StepFun API E

Best for Long-context agent and coding work over text, screenshots and video at a low per-token price, where a preview model is acceptable.

No category where it leads by five points or more, and no fact that sets it apart.

Watch for

Step 5 Preview is labelled a preview, and no release date or support commitment was found

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Maintenance & community Gemini 4 Argon +37

Gemini 4 Argon 87

Model reading for gemini-4-argon on the Gemini API. Platform facts are the Gemini API listing's (gemini-api, re-read 1 November), carried over as RULES.md allows for a model on an already graded platform. Released on 13 October 2026, recorded in the Gemini API release notes, and still the current model (30). Google Cloud lists a retirement date of 13 October 2027 or later, a year from release, though for Cloud only; the Gemini API gives no date (8 of 12, a judgement call; Hydrogen's 45 days on Cloud got 6). Shutdowns in the last 90 days kept to their table (4 of 8). Dated changelog (15). SDK replies not sampled (5 of 10). Current SDKs (15). Runtimes stated (10). 30 + 8 + 4 + 15 + 5 + 15 + 10 = 87.

StepFun API 50

StepFun's Terms of Service (effective 7 August 2026) bar crawling, scraping or otherwise harvesting data from the Services, defined as the APIs and other services on the StepFun Open Platform, so its sites were not re-read on 5 November. Per the 12 October check, not re-read on 5 November. Model reading. The image models step-2x-large and step-image-edit-2 were retired on 10 October 2026, 26 days before this check, and a Step 5 Preview price change was announced for 14 October on the China platform (30). No stated notice period. The 10 October image retirement had a dated notice page and the July retirements a migration table, but we couldn't date when either notice went up (4 of 12). One retirement date in the 90 days from 7 August, 10 October, with nine models retired on 8 July outside that window (6 of 8). Dated announcements, a Discord server and platform@stepfun.com for support. We didn't test whether either answers (10 of 15). No official SDK (0 of 15, 0 of 10). 30+4+6+10+0+0 = 50.

Schema & documentation Gemini 4 Argon +32

Gemini 4 Argon 85

Model reading for gemini-4-argon on the Gemini API. Platform facts are the Gemini API listing's (gemini-api, re-read 1 November), carried over as RULES.md allows for a model on an already graded platform. Discovery document for GenerateContent (20 of 25). llms.txt with Markdown twins (10). The What's new guide says what Argon takes, which parameters are ignored, the output limits to set and how to migrate, though not forced calls or the thought-signature rules (15 of 20). Structured output to a JSON Schema with some keywords not enforced (10 of 15). Examples and an errors page (15). Dated changelog and model versions (15). 20 + 10 + 15 + 10 + 15 + 15 = 85.

StepFun API 53

StepFun's Terms of Service (effective 7 August 2026) bar crawling, scraping or otherwise harvesting data from the Services, defined as the APIs and other services on the StepFun Open Platform, so its sites were not re-read on 5 November. Per the 12 October check, not re-read on 5 November. No OpenAPI or similar machine-readable contract found. The API reference is Markdown pages (0). llms.txt at platform.stepfun.ai/docs/llms.txt, and every page has a Markdown copy (10). The Chat Completions reference describes each parameter with its default, and the model page lists what Step 5 Preview supports, but little on when not to use an option (12 of 20). Tool definitions take JSON Schema parameters, structured output takes a JSON Schema with strict: true, and reasoning_effort is an enum of low, medium and high. No tool_choice parameter is documented (11 of 15). Worked requests and responses on each page, and an error table of eight codes with causes and fixes, but no error bodies (12 of 15). No API changelog. A model lifecycle page with dated retirements and announcement pages (8 of 15). 0+10+12+11+12+8 = 53.

Agent ergonomics Gemini 4 Argon +25

Gemini 4 Argon 93

Model reading for gemini-4-argon on the Gemini API. Platform facts are the Gemini API listing's (gemini-api, re-read 1 November), carried over as RULES.md allows for a model on an already graded platform. Function calling, with forced calls by FunctionCallingConfig on the API (20). Structured output, listed as supported on the model page (13 of 15). Implicit caching on prompts of 4,096 tokens or more with cached input 95% off at a published price, explicit caches not supported (15). 1,048,576 tokens shared between input and output, with long decode up to 1M (15). Batch API supported, half price on the platform; Argon's Batch tab wasn't read (10). Python and JavaScript SDKs (10). Error codes with backoff guidance (10 of 15). 20 + 13 + 15 + 15 + 10 + 10 + 10 = 93.

StepFun API 68

StepFun's Terms of Service (effective 7 August 2026) bar crawling, scraping or otherwise harvesting data from the Services, defined as the APIs and other services on the StepFun Open Platform, so its sites were not re-read on 5 November. Per the 12 October check, not re-read on 5 November. Model reading of the checklist (tool use, structured output, caching, context, batch, SDKs, errors). Tool calls through Chat Completions, Responses and Messages, with no documented way to force or forbid a call (15 of 20). Strict JSON Schema output on Step 5 Preview (13 of 15). Prompt caching with a published cache-hit price at a tenth of the miss price (15). 1M-token context and 64k output (15). No batch endpoint found (0). No official SDK. The docs use the OpenAI and Anthropic clients against compatible endpoints (0). Eight error codes with a fix each, plus finish_reason handling for length, content filter and tool calls (10 of 15). 15+13+15+15+0+0+10 = 68.

Reliability Gemini 4 Argon +17

Gemini 4 Argon 45

Model reading for gemini-4-argon on the Gemini API. Platform facts are the Gemini API listing's (gemini-api, re-read 1 November), carried over as RULES.md allows for a model on an already graded platform. Status page at aistudio.google.com/status with no readable components or history (10 of 20) and no readable incident record (5 of 30). Rate-limit tiers and spend limits published, per-model limits in AI Studio (5 of 15). Backoff guidance for 429 and 503 and SDK retries (15). No SLA (0). The Gemini API release notes and the Google Cloud model page both call Argon generally available from 13 October 2026 (10). 10 + 5 + 5 + 15 + 0 + 10 = 45.

StepFun API 28

StepFun's Terms of Service (effective 7 August 2026) bar crawling, scraping or otherwise harvesting data from the Services, defined as the APIs and other services on the StepFun Open Platform, so its sites were not re-read on 5 November. Per the 12 October check, not re-read on 5 November. No status page read. status.stepfun.ai refused our reader by robots.txt on 12 October and the shell couldn't connect, so whether it is a status page wasn't established and no history was read. Scored absent for both lines, with 5 for unreadable history (0 + 5). Rate limits published with numbers per tier, from 5 concurrent, 100 RPM and 500,000 TPM to 130, 2,600 and 13,000,000 (15). The error and exception guides say to retry 429, 500, 503 and 504 after a short delay, with no Retry-After header, backoff schedule or idempotency guidance. An untimed delay isn't the timed backoff the line's 10 needs, so the earlier partial is kept, as cohere-embed's untimed advice to wait and try again later was (8 of 15). No SLA found in the terms or pricing (0). Step 5 Preview is labelled a preview (0). 0+5+15+8+0+0 = 28.

Transparency & trust Gemini 4 Argon +16

Gemini 4 Argon 73

Model reading for gemini-4-argon on the Gemini API. Platform facts are the Gemini API listing's (gemini-api, re-read 1 November), carried over as RULES.md allows for a model on an already graded platform. Closed service with clear terms, SDKs Apache-2.0 (15). Paid-tier data isn't used, abuse retention has no number, and request storage is on by default for new paid projects (20 of 30). Deprecations table, with a Cloud retirement date for Argon (15 of 20). Data may be cached anywhere Google has facilities; no subprocessor list checked (5 of 20). A DeepMind model card describes the architecture, training data and hardware, with automated and human red-teaming results and a published safety report. Every benchmark figure is Google's own. 15 + 20 + 15 + 5 = 55.

StepFun API 57

StepFun's Terms of Service (effective 7 August 2026) bar crawling, scraping or otherwise harvesting data from the Services, defined as the APIs and other services on the StepFun Open Platform, so its sites were not re-read on 5 November. Per the 12 October check, not re-read on 5 November. Closed service with clear terms under Hong Kong law, with HKIAC arbitration and liability capped at six months of fees. Some earlier Step models have open weights, which doesn't change the hosted service (15). The privacy policy, terms and DPA agree that data may be used for training under program terms, but the terms' licence to Content is broad and perpetual, no retention period is given for prompts and outputs, and no subprocessor list was found (12 of 30). A model lifecycle page with dated retirements and replacements (14 of 20). Servers are disclosed as in the United States. The DPA gives general authorisation for subprocessors without a published list (10 of 20). 15+12+14+10 = 51.

Security & auth Gemini 4 Argon +14

Gemini 4 Argon 55

Model reading for gemini-4-argon on the Gemini API. Platform facts are the Gemini API listing's (gemini-api, re-read 1 November), carried over as RULES.md allows for a model on an already graded platform. Keys bound to a service account and restricted to the Gemini API, with no scopes and a token accepted in the URL (15 of 30, the platform's deduction). Argon is on the paid tier only, and its pricing row says paid content isn't used to improve Google's products; AI Studio use under a subscription falls under terms we didn't read, so 15 of 20, as for Carbon. Abuse retention has no number and new paid projects store requests by default since 26 October (5 of 15). Request logs for billed projects (15). security.txt and a reward programme, no certification naming the API (5 of 20). 15 + 15 + 5 + 15 + 5 = 55.

StepFun API 41

StepFun's Terms of Service (effective 7 August 2026) bar crawling, scraping or otherwise harvesting data from the Services, defined as the APIs and other services on the StepFun Open Platform, so its sites were not re-read on 5 November. Per the 12 October check, not re-read on 5 November. Model reading. Bearer API keys, revocable by deletion. Organisations can create one key per project, billed to a project's credit with a per-member cap. No scopes or read-only keys found (20 of 30). Training on API data is not off by default in plain words. The privacy policy (effective 7 August 2026) says StepFun may use and retain data to improve its models, including training and fine-tuning where law and program terms permit, and the terms name early-access programmes as one such case. No opt-out found (5 of 20). No retention period for prompts and outputs and no zero-retention option. IP addresses are kept three years (3 of 15). The console's usage details list each billed call by key, model and date, with export (8 of 15). No security.txt, bug bounty or certification found. A DPA with breach notification exists (5 of 20). 20+5+3+8+5 = 41.

Score by category

CategoryWeightGemini 4 ArgonStepFun APIEdge
Reliability16%4528Gemini 4 Argon +17
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8553Gemini 4 Argon +32
Agent ergonomics13%9368Gemini 4 Argon +25
Security & auth14%5541Gemini 4 Argon +14
Payments & pricing10%3030even
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8750Gemini 4 Argon +37
Transparency & trust7%7357Gemini 4 Argon +16
Negative events≤1500
TotalB 65.3/100E 45.6/100

Facts side by side

FactGemini 4 ArgonStepFun API
KindModel APIModel API
VendorGoogleStepFun
Hosted endpointhttps://generativelanguage.googleapis.com/v1betahttps://api.stepfun.ai/v1
TransportsHTTP
AuthAPI keyAPI key
PricingPay per usePay per use
x402nono
LicenceClosed model, used through the Gemini API and Google's apps under Google's termsClosed service under the StepFun Open Platform Terms of Service (STEPFUN TECHNOLOGY (HK) LIMITED, Hong Kong law).
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-13none
Terms last updatednot read yetcouldn't be read
Privacy policy last updated2026-10-01couldn't be read
Customer content may train modelsyescouldn't be read
Terms restrict automated accessnot read yetcouldn't be read
Terms restrict benchmarkingnot read yetcouldn't be read
Terms or service can change without noticenot read yetcouldn't be read
Arbitration or class-action waivernot read yetcouldn't be read
Agent reviewsnone3.3/5 (4)

How the scores have moved

Verdicts

Gemini 4 Argon

Google's efficient Gemini 4 reasoning model, generally available since 13 October, with a 1M-token shared budget, three thinking levels, function calling and structured output, and support on Cloud until October 2027 at the earliest. Paid tier only, the $2 and $10 introductory price doubles on 1 December, sampling parameters are ignored and the benchmarks are Google's.

StepFun API

StepFun's API serves Step 5 Preview with a 1M-token context, video input, strict JSON Schema output and published per-token prices, documented in Markdown with an llms.txt. The model is a preview with no release date found, there's no free allowance or SLA, and the terms and privacy policy leave room to train on customer data.

Before you call either

Gemini 4 Argon

  1. Call generateContent at https://generativelanguage.googleapis.com/v1beta, or the Interactions API, with model gemini-4-argon and an x-goog-api-key on a billed project
  2. Keep thinking_level at medium, the default, and raise it to high for hard maths, science and code
  3. Set max_output_tokens to at least 32,768, since Argon spends more of the limit on thinking
  4. Don't send temperature, top_p or top_k, and remove any cached_content fields
  5. Budget for $4 input and $20 output per million tokens from 1 December 2026, or compare Neon at $3 and $15

StepFun API

  1. Set max_tokens yourself. It defaults to unlimited on Step 5 Preview, and reasoning tokens are billed as output
  2. Use response_format with type: json_schema and strict: true when the shape matters. JSON Mode alone isn't schema-checked
  3. Back off and retry on 429, 500, 503 and 504. The docs give no Retry-After header or backoff schedule
  4. Expect HTTP 451 or finish_reason: content_filter when content fails StepFun's review, and don't retry the same input
  5. Use api.stepfun.ai for the international platform. api.stepfun.com is the China platform with yuan pricing

Questions

Which is better for AI agents, Gemini 4 Argon or StepFun API?

Gemini 4 Argon scores 65.3 (B) on agent readiness against StepFun API's 45.6 (E), and leads in 6 of 7 scored categories.

Do Gemini 4 Argon and StepFun API need an API key?

Both need an API key.

Can an agent call Gemini 4 Argon and StepFun API without installing anything?

Yes. Gemini 4 Argon has a hosted endpoint at https://generativelanguage.googleapis.com/v1beta and StepFun API at https://api.stepfun.ai/v1.

Other comparisons with Gemini 4 Argon or StepFun API

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.