PulseAugur
EN
LIVE 10:47:32

AI model batch inference costs plummet, some by 67% in a month · 8 sources tracked

By PulseAugur Editorial · [9 sources] ·

Several leading AI models have seen significant price reductions in their batch inference costs over the past month, with some dropping by as much as 67%. Models like DeepSeek V4.1 Flash, GPT-5.6 Sol Pro, Mistral Large 3 2512, and GLM 5.3 Flash are now considerably cheaper to use, with some also offering open weights. These price drops suggest a rapid decrease in the cost of running large language models, potentially driven by advancements in inference efficiency and hardware. AI

IMPACT Accelerates adoption of advanced LLMs by reducing operational costs and potentially enabling new use cases.

RANK_REASON Multiple AI models from different providers have seen significant price drops in batch inference costs, indicating a broader industry trend.

Read on Mastodon — mastodon.social →

AI-generated summary · Google Gemini · from 9 sources. How we write summaries →

AI model batch inference costs plummet, some by 67% in a month · 8 sources tracked

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Significant
Multiple AI models from different providers have seen significant price drops in batch inference costs, indicating a broader industry trend.
Source corroboration
9 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
16 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+1 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

Full methodology in our editorial standards.

COVERAGE [9]

  1. Mastodon — sigmoid.social TIER_1 English(EN) · [email protected] ·

    DeepSeek V4.1 Flash: 1M-token context, open weights, $0.15 in / $0.6 out per 1M — long-context work without the usual price wall. Hourly-updated model data on o

    DeepSeek V4.1 Flash: 1M-token context, open weights, $0.15 in / $0.6 out per 1M — long-context work without the usual price wall. Hourly-updated model data on olud.ai: https:// olud.ai/latest.html # AI # LLM # OpenSource

    LINKS olud.ai/latest.html
  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    GPT-5.6 Sol Pro’s batch pricing just fell 67% in a month — from $15 to $5 per 1M output tokens. That’s a signal batch inference costs are collapsing faster than

    GPT-5.6 Sol Pro’s batch pricing just fell 67% in a month — from $15 to $5 per 1M output tokens. That’s a signal batch inference costs are collapsing faster than most budgets assumed. See what’s driving the drop on olud.ai. https:// olud.ai/model/gpt-5-6-sol-pro- batch.html # AI #…

    LINKS olud.ai/…/gpt-5-6-sol-pro-batch.html
  3. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Batch price for Mistral Large 3 2512 dropped 50% in 30 days, to $0.75 per 1M output tokens—and it’s open weights. https:// olud.ai/pricing.html # AI # LLM # Pri

    Batch price for Mistral Large 3 2512 dropped 50% in 30 days, to $0.75 per 1M output tokens—and it’s open weights. https:// olud.ai/pricing.html # AI # LLM # Pricing

    LINKS olud.ai/pricing.html
  4. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    Mistral Small 4 (batch) dropped to $0.3 per 1M output tokens — a 50% cut in 30 days, and the weights are open. That kind of price movement on an open model is r

    Mistral Small 4 (batch) dropped to $0.3 per 1M output tokens — a 50% cut in 30 days, and the weights are open. That kind of price movement on an open model is rare. https:// olud.ai/pricing.html # AI # LLM # Pricing

    LINKS olud.ai/pricing.html
  5. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📉 GLM 5.3 Flash (batch) just halved its output price to $0.25/1M tokens — a 50% drop in 30 days, and it’s open weights. https:// olud.ai/pricing.html # AI # LLM

    📉 GLM 5.3 Flash (batch) just halved its output price to $0.25/1M tokens — a 50% drop in 30 days, and it’s open weights. https:// olud.ai/pricing.html # AI # LLM # Pricing

    LINKS olud.ai/pricing.html
  6. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    DeepSeek V4 Flash 0423 output tokens now cost $0.13 per million, half of what they did a month ago. Open weights make the drop even harder to ignore. https:// o

    DeepSeek V4 Flash 0423 output tokens now cost $0.13 per million, half of what they did a month ago. Open weights make the drop even harder to ignore. https:// olud.ai/model/deepseek-v4-flas h-0423.html # AI # LLM # Pricing

    LINKS olud.ai/…/deepseek-v4-flash-0423.html
  7. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📉 GPT-5.6 Sol (batch) dropped 67% in 30 days—from $15 to $5 per 1M output tokens. Batch pricing is now a quarter of its cost last month. See the full trend on o

    📉 GPT-5.6 Sol (batch) dropped 67% in 30 days—from $15 to $5 per 1M output tokens. Batch pricing is now a quarter of its cost last month. See the full trend on olud.ai. https:// olud.ai/model/gpt-5-6-sol-batc h.html # AI # LLM # Pricing

    LINKS olud.ai/…/gpt-5-6-sol-batch.html
  8. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    GPT-5.6 Sol Pro dropped 67% in 30 days — $30 to $10 per 1M output tokens. That kind of speed usually means the market is moving fast. See the data on olud.ai. h

    GPT-5.6 Sol Pro dropped 67% in 30 days — $30 to $10 per 1M output tokens. That kind of speed usually means the market is moving fast. See the data on olud.ai. https:// olud.ai/model/gpt-5-6-sol-pro. html # AI # LLM # Pricing

    LINKS olud.ai/…/gpt-5-6-sol-pro.html
  9. Mastodon — mastodon.social TIER_1 English(EN) · opensourceaitech ·

    📉 GPT-5.6 Sol dropped 67% in 30 days, now $10 per 1M output tokens. What else has slid this fast? https:// olud.ai/model/gpt-5-6-sol.html # AI # LLM # Pricing

    📉 GPT-5.6 Sol dropped 67% in 30 days, now $10 per 1M output tokens. What else has slid this fast? https:// olud.ai/model/gpt-5-6-sol.html # AI # LLM # Pricing

    LINKS olud.ai/…/gpt-5-6-sol.html

RELATED ENTITIES

RELATED TOPICS